// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
// REQUIRES: aarch64-registered-target
// RUN: %clang_cc1 -triple aarch64 -target-feature +sme -target-feature +sme2 -O1 -Werror -Wall -emit-llvm -o - %s | FileCheck %s
// RUN: %clang_cc1 -triple aarch64 -target-feature +sve -target-feature +sve2 -target-feature +sve2p1 -O1 -Werror -Wall -emit-llvm -o - -x c++ %s | FileCheck %s -check-prefix=CPP-CHECK
// RUN: %clang_cc1 -DSVE_OVERLOADED_FORMS -triple aarch64 -target-feature +sve -target-feature +sve2 -target-feature +sve2p1 -O1 -Werror -Wall -emit-llvm -o - %s | FileCheck %s
// RUN: %clang_cc1 -DSVE_OVERLOADED_FORMS -triple aarch64 -target-feature +sve -target-feature +sve2 -target-feature +sve2p1 -O1 -Werror -Wall -emit-llvm -o - -x c++ %s | FileCheck %s -check-prefix=CPP-CHECK
// RUN: %clang_cc1 -triple aarch64 -target-feature +sve -target-feature +sve2 -target-feature +sve2p1 -S -disable-O0-optnone -Werror -Wall -o /dev/null %s
// RUN: %clang_cc1 -triple aarch64 -target-feature +sme -target-feature +sme2 -S -disable-O0-optnone -Werror -Wall -o /dev/null %s

#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

#ifdef __ARM_FEATURE_SME
#define ATTR __arm_streaming
#else
#define ATTR
#endif

// CHECK-LABEL: @test_svstnt1_u8_x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv16i8(<vscale x 16 x i8> [[V_COERCE0:%.*]], <vscale x 16 x i8> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z18test_svstnt1_u8_x2u11__SVCount_tPh11svuint8x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv16i8(<vscale x 16 x i8> [[V_COERCE0:%.*]], <vscale x 16 x i8> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_u8_x2(svcount_t pn, uint8_t *base, svuint8x2_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1,_u8_x2,,)(pn, base, v);
}


// CHECK-LABEL: @test_svstnt1_u16_x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv8i16(<vscale x 8 x i16> [[V_COERCE0:%.*]], <vscale x 8 x i16> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z19test_svstnt1_u16_x2u11__SVCount_tPt12svuint16x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv8i16(<vscale x 8 x i16> [[V_COERCE0:%.*]], <vscale x 8 x i16> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_u16_x2(svcount_t pn, uint16_t *base, svuint16x2_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1,_u16_x2,,)(pn, base, v);
}


// CHECK-LABEL: @test_svstnt1_u32_x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv4i32(<vscale x 4 x i32> [[V_COERCE0:%.*]], <vscale x 4 x i32> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z19test_svstnt1_u32_x2u11__SVCount_tPj12svuint32x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv4i32(<vscale x 4 x i32> [[V_COERCE0:%.*]], <vscale x 4 x i32> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_u32_x2(svcount_t pn, uint32_t *base, svuint32x2_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1,_u32_x2,,)(pn, base, v);
}


// CHECK-LABEL: @test_svstnt1_u64_x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv2i64(<vscale x 2 x i64> [[V_COERCE0:%.*]], <vscale x 2 x i64> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z19test_svstnt1_u64_x2u11__SVCount_tPm12svuint64x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv2i64(<vscale x 2 x i64> [[V_COERCE0:%.*]], <vscale x 2 x i64> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_u64_x2(svcount_t pn, uint64_t *base, svuint64x2_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1,_u64_x2,,)(pn, base, v);
}


// CHECK-LABEL: @test_svstnt1_u8_x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv16i8(<vscale x 16 x i8> [[V_COERCE0:%.*]], <vscale x 16 x i8> [[V_COERCE1:%.*]], <vscale x 16 x i8> [[V_COERCE2:%.*]], <vscale x 16 x i8> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z18test_svstnt1_u8_x4u11__SVCount_tPh11svuint8x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv16i8(<vscale x 16 x i8> [[V_COERCE0:%.*]], <vscale x 16 x i8> [[V_COERCE1:%.*]], <vscale x 16 x i8> [[V_COERCE2:%.*]], <vscale x 16 x i8> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_u8_x4(svcount_t pn, uint8_t *base, svuint8x4_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1,_u8_x4,,)(pn, base, v);
}


// CHECK-LABEL: @test_svstnt1_u16_x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv8i16(<vscale x 8 x i16> [[V_COERCE0:%.*]], <vscale x 8 x i16> [[V_COERCE1:%.*]], <vscale x 8 x i16> [[V_COERCE2:%.*]], <vscale x 8 x i16> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z19test_svstnt1_u16_x4u11__SVCount_tPt12svuint16x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv8i16(<vscale x 8 x i16> [[V_COERCE0:%.*]], <vscale x 8 x i16> [[V_COERCE1:%.*]], <vscale x 8 x i16> [[V_COERCE2:%.*]], <vscale x 8 x i16> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_u16_x4(svcount_t pn, uint16_t *base, svuint16x4_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1,_u16_x4,,)(pn, base, v);
}


// CHECK-LABEL: @test_svstnt1_u32_x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv4i32(<vscale x 4 x i32> [[V_COERCE0:%.*]], <vscale x 4 x i32> [[V_COERCE1:%.*]], <vscale x 4 x i32> [[V_COERCE2:%.*]], <vscale x 4 x i32> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z19test_svstnt1_u32_x4u11__SVCount_tPj12svuint32x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv4i32(<vscale x 4 x i32> [[V_COERCE0:%.*]], <vscale x 4 x i32> [[V_COERCE1:%.*]], <vscale x 4 x i32> [[V_COERCE2:%.*]], <vscale x 4 x i32> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_u32_x4(svcount_t pn, uint32_t *base, svuint32x4_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1,_u32_x4,,)(pn, base, v);
}


// CHECK-LABEL: @test_svstnt1_u64_x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv2i64(<vscale x 2 x i64> [[V_COERCE0:%.*]], <vscale x 2 x i64> [[V_COERCE1:%.*]], <vscale x 2 x i64> [[V_COERCE2:%.*]], <vscale x 2 x i64> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z19test_svstnt1_u64_x4u11__SVCount_tPm12svuint64x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv2i64(<vscale x 2 x i64> [[V_COERCE0:%.*]], <vscale x 2 x i64> [[V_COERCE1:%.*]], <vscale x 2 x i64> [[V_COERCE2:%.*]], <vscale x 2 x i64> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_u64_x4(svcount_t pn, uint64_t *base, svuint64x4_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1,_u64_x4,,)(pn, base, v);
}


// CHECK-LABEL: @test_svstnt1_s8_x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv16i8(<vscale x 16 x i8> [[V_COERCE0:%.*]], <vscale x 16 x i8> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z18test_svstnt1_s8_x2u11__SVCount_tPa10svint8x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv16i8(<vscale x 16 x i8> [[V_COERCE0:%.*]], <vscale x 16 x i8> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_s8_x2(svcount_t pn, int8_t *base, svint8x2_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1,_s8_x2,,)(pn, base, v);
}


// CHECK-LABEL: @test_svstnt1_s16_x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv8i16(<vscale x 8 x i16> [[V_COERCE0:%.*]], <vscale x 8 x i16> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z19test_svstnt1_s16_x2u11__SVCount_tPs11svint16x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv8i16(<vscale x 8 x i16> [[V_COERCE0:%.*]], <vscale x 8 x i16> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_s16_x2(svcount_t pn, int16_t *base, svint16x2_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1,_s16_x2,,)(pn, base, v);
}


// CHECK-LABEL: @test_svstnt1_s32_x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv4i32(<vscale x 4 x i32> [[V_COERCE0:%.*]], <vscale x 4 x i32> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z19test_svstnt1_s32_x2u11__SVCount_tPi11svint32x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv4i32(<vscale x 4 x i32> [[V_COERCE0:%.*]], <vscale x 4 x i32> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_s32_x2(svcount_t pn, int32_t *base, svint32x2_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1,_s32_x2,,)(pn, base, v);
}


// CHECK-LABEL: @test_svstnt1_s64_x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv2i64(<vscale x 2 x i64> [[V_COERCE0:%.*]], <vscale x 2 x i64> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z19test_svstnt1_s64_x2u11__SVCount_tPl11svint64x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv2i64(<vscale x 2 x i64> [[V_COERCE0:%.*]], <vscale x 2 x i64> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_s64_x2(svcount_t pn, int64_t *base, svint64x2_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1,_s64_x2,,)(pn, base, v);
}


// CHECK-LABEL: @test_svstnt1_s8_x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv16i8(<vscale x 16 x i8> [[V_COERCE0:%.*]], <vscale x 16 x i8> [[V_COERCE1:%.*]], <vscale x 16 x i8> [[V_COERCE2:%.*]], <vscale x 16 x i8> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z18test_svstnt1_s8_x4u11__SVCount_tPa10svint8x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv16i8(<vscale x 16 x i8> [[V_COERCE0:%.*]], <vscale x 16 x i8> [[V_COERCE1:%.*]], <vscale x 16 x i8> [[V_COERCE2:%.*]], <vscale x 16 x i8> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_s8_x4(svcount_t pn, int8_t *base, svint8x4_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1,_s8_x4,,)(pn, base, v);
}


// CHECK-LABEL: @test_svstnt1_s16_x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv8i16(<vscale x 8 x i16> [[V_COERCE0:%.*]], <vscale x 8 x i16> [[V_COERCE1:%.*]], <vscale x 8 x i16> [[V_COERCE2:%.*]], <vscale x 8 x i16> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z19test_svstnt1_s16_x4u11__SVCount_tPs11svint16x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv8i16(<vscale x 8 x i16> [[V_COERCE0:%.*]], <vscale x 8 x i16> [[V_COERCE1:%.*]], <vscale x 8 x i16> [[V_COERCE2:%.*]], <vscale x 8 x i16> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_s16_x4(svcount_t pn, int16_t *base, svint16x4_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1,_s16_x4,,)(pn, base, v);
}


// CHECK-LABEL: @test_svstnt1_s32_x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv4i32(<vscale x 4 x i32> [[V_COERCE0:%.*]], <vscale x 4 x i32> [[V_COERCE1:%.*]], <vscale x 4 x i32> [[V_COERCE2:%.*]], <vscale x 4 x i32> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z19test_svstnt1_s32_x4u11__SVCount_tPi11svint32x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv4i32(<vscale x 4 x i32> [[V_COERCE0:%.*]], <vscale x 4 x i32> [[V_COERCE1:%.*]], <vscale x 4 x i32> [[V_COERCE2:%.*]], <vscale x 4 x i32> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_s32_x4(svcount_t pn, int32_t *base, svint32x4_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1,_s32_x4,,)(pn, base, v);
}


// CHECK-LABEL: @test_svstnt1_s64_x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv2i64(<vscale x 2 x i64> [[V_COERCE0:%.*]], <vscale x 2 x i64> [[V_COERCE1:%.*]], <vscale x 2 x i64> [[V_COERCE2:%.*]], <vscale x 2 x i64> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z19test_svstnt1_s64_x4u11__SVCount_tPl11svint64x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv2i64(<vscale x 2 x i64> [[V_COERCE0:%.*]], <vscale x 2 x i64> [[V_COERCE1:%.*]], <vscale x 2 x i64> [[V_COERCE2:%.*]], <vscale x 2 x i64> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_s64_x4(svcount_t pn, int64_t *base, svint64x4_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1,_s64_x4,,)(pn, base, v);
}


// CHECK-LABEL: @test_svstnt1_f16_x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv8f16(<vscale x 8 x half> [[V_COERCE0:%.*]], <vscale x 8 x half> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z19test_svstnt1_f16_x2u11__SVCount_tPDh13svfloat16x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv8f16(<vscale x 8 x half> [[V_COERCE0:%.*]], <vscale x 8 x half> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_f16_x2(svcount_t pn, float16_t *base, svfloat16x2_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1,_f16_x2,,)(pn, base, v);
}


// CHECK-LABEL: @test_svstnt1_f32_x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv4f32(<vscale x 4 x float> [[V_COERCE0:%.*]], <vscale x 4 x float> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z19test_svstnt1_f32_x2u11__SVCount_tPf13svfloat32x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv4f32(<vscale x 4 x float> [[V_COERCE0:%.*]], <vscale x 4 x float> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_f32_x2(svcount_t pn, float32_t *base, svfloat32x2_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1,_f32_x2,,)(pn, base, v);
}


// CHECK-LABEL: @test_svstnt1_f64_x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv2f64(<vscale x 2 x double> [[V_COERCE0:%.*]], <vscale x 2 x double> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z19test_svstnt1_f64_x2u11__SVCount_tPd13svfloat64x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv2f64(<vscale x 2 x double> [[V_COERCE0:%.*]], <vscale x 2 x double> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_f64_x2(svcount_t pn, float64_t *base, svfloat64x2_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1,_f64_x2,,)(pn, base, v);
}

// CHECK-LABEL: @test_svstnt1_mf8_x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv16i8(<vscale x 16 x i8> [[V_COERCE0:%.*]], <vscale x 16 x i8> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z19test_svstnt1_mf8_x2u11__SVCount_tPu6__mfp813svmfloat8x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv16i8(<vscale x 16 x i8> [[V_COERCE0:%.*]], <vscale x 16 x i8> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_mf8_x2(svcount_t pn, mfloat8_t *base, svmfloat8x2_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1,_mf8_x2,,)(pn, base, v);
}


// CHECK-LABEL: @test_svstnt1_f16_x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv8f16(<vscale x 8 x half> [[V_COERCE0:%.*]], <vscale x 8 x half> [[V_COERCE1:%.*]], <vscale x 8 x half> [[V_COERCE2:%.*]], <vscale x 8 x half> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z19test_svstnt1_f16_x4u11__SVCount_tPDh13svfloat16x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv8f16(<vscale x 8 x half> [[V_COERCE0:%.*]], <vscale x 8 x half> [[V_COERCE1:%.*]], <vscale x 8 x half> [[V_COERCE2:%.*]], <vscale x 8 x half> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_f16_x4(svcount_t pn, float16_t *base, svfloat16x4_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1,_f16_x4,,)(pn, base, v);
}


// CHECK-LABEL: @test_svstnt1_f32_x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv4f32(<vscale x 4 x float> [[V_COERCE0:%.*]], <vscale x 4 x float> [[V_COERCE1:%.*]], <vscale x 4 x float> [[V_COERCE2:%.*]], <vscale x 4 x float> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z19test_svstnt1_f32_x4u11__SVCount_tPf13svfloat32x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv4f32(<vscale x 4 x float> [[V_COERCE0:%.*]], <vscale x 4 x float> [[V_COERCE1:%.*]], <vscale x 4 x float> [[V_COERCE2:%.*]], <vscale x 4 x float> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_f32_x4(svcount_t pn, float32_t *base, svfloat32x4_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1,_f32_x4,,)(pn, base, v);
}


// CHECK-LABEL: @test_svstnt1_f64_x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv2f64(<vscale x 2 x double> [[V_COERCE0:%.*]], <vscale x 2 x double> [[V_COERCE1:%.*]], <vscale x 2 x double> [[V_COERCE2:%.*]], <vscale x 2 x double> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z19test_svstnt1_f64_x4u11__SVCount_tPd13svfloat64x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv2f64(<vscale x 2 x double> [[V_COERCE0:%.*]], <vscale x 2 x double> [[V_COERCE1:%.*]], <vscale x 2 x double> [[V_COERCE2:%.*]], <vscale x 2 x double> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_f64_x4(svcount_t pn, float64_t *base, svfloat64x4_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1,_f64_x4,,)(pn, base, v);
}

// CHECK-LABEL: @test_svstnt1_mf8_x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv16i8(<vscale x 16 x i8> [[V_COERCE0:%.*]], <vscale x 16 x i8> [[V_COERCE1:%.*]], <vscale x 16 x i8> [[V_COERCE2:%.*]], <vscale x 16 x i8> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z19test_svstnt1_mf8_x4u11__SVCount_tPu6__mfp813svmfloat8x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv16i8(<vscale x 16 x i8> [[V_COERCE0:%.*]], <vscale x 16 x i8> [[V_COERCE1:%.*]], <vscale x 16 x i8> [[V_COERCE2:%.*]], <vscale x 16 x i8> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_mf8_x4(svcount_t pn, mfloat8_t *base, svmfloat8x4_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1,_mf8_x4,,)(pn, base, v);
}

// == VNUM variants ==


// CHECK-LABEL: @test_svstnt1_vnum_u8_x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv16i8(<vscale x 16 x i8> [[V_COERCE0:%.*]], <vscale x 16 x i8> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z23test_svstnt1_vnum_u8_x2u11__SVCount_tPhl11svuint8x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CPP-CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv16i8(<vscale x 16 x i8> [[V_COERCE0:%.*]], <vscale x 16 x i8> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_u8_x2(svcount_t pn, uint8_t *base, int64_t vnum, svuint8x2_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_u8_x2,,)(pn, base, vnum, v);
}


// CHECK-LABEL: @test_svstnt1_vnum_u16_x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv8i16(<vscale x 8 x i16> [[V_COERCE0:%.*]], <vscale x 8 x i16> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z24test_svstnt1_vnum_u16_x2u11__SVCount_tPtl12svuint16x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CPP-CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv8i16(<vscale x 8 x i16> [[V_COERCE0:%.*]], <vscale x 8 x i16> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_u16_x2(svcount_t pn, uint16_t *base, int64_t vnum, svuint16x2_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_u16_x2,,)(pn, base, vnum, v);
}


// CHECK-LABEL: @test_svstnt1_vnum_u32_x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv4i32(<vscale x 4 x i32> [[V_COERCE0:%.*]], <vscale x 4 x i32> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z24test_svstnt1_vnum_u32_x2u11__SVCount_tPjl12svuint32x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CPP-CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv4i32(<vscale x 4 x i32> [[V_COERCE0:%.*]], <vscale x 4 x i32> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_u32_x2(svcount_t pn, uint32_t *base, int64_t vnum, svuint32x2_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_u32_x2,,)(pn, base, vnum, v);
}


// CHECK-LABEL: @test_svstnt1_vnum_u64_x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv2i64(<vscale x 2 x i64> [[V_COERCE0:%.*]], <vscale x 2 x i64> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z24test_svstnt1_vnum_u64_x2u11__SVCount_tPml12svuint64x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CPP-CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv2i64(<vscale x 2 x i64> [[V_COERCE0:%.*]], <vscale x 2 x i64> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_u64_x2(svcount_t pn, uint64_t *base, int64_t vnum, svuint64x2_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_u64_x2,,)(pn, base, vnum, v);
}


// CHECK-LABEL: @test_svstnt1_vnum_u8_x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv16i8(<vscale x 16 x i8> [[V_COERCE0:%.*]], <vscale x 16 x i8> [[V_COERCE1:%.*]], <vscale x 16 x i8> [[V_COERCE2:%.*]], <vscale x 16 x i8> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z23test_svstnt1_vnum_u8_x4u11__SVCount_tPhl11svuint8x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CPP-CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv16i8(<vscale x 16 x i8> [[V_COERCE0:%.*]], <vscale x 16 x i8> [[V_COERCE1:%.*]], <vscale x 16 x i8> [[V_COERCE2:%.*]], <vscale x 16 x i8> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_u8_x4(svcount_t pn, uint8_t *base, int64_t vnum, svuint8x4_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_u8_x4,,)(pn, base, vnum, v);
}


// CHECK-LABEL: @test_svstnt1_vnum_u16_x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv8i16(<vscale x 8 x i16> [[V_COERCE0:%.*]], <vscale x 8 x i16> [[V_COERCE1:%.*]], <vscale x 8 x i16> [[V_COERCE2:%.*]], <vscale x 8 x i16> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z24test_svstnt1_vnum_u16_x4u11__SVCount_tPtl12svuint16x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CPP-CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv8i16(<vscale x 8 x i16> [[V_COERCE0:%.*]], <vscale x 8 x i16> [[V_COERCE1:%.*]], <vscale x 8 x i16> [[V_COERCE2:%.*]], <vscale x 8 x i16> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_u16_x4(svcount_t pn, uint16_t *base, int64_t vnum, svuint16x4_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_u16_x4,,)(pn, base, vnum, v);
}


// CHECK-LABEL: @test_svstnt1_vnum_u32_x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv4i32(<vscale x 4 x i32> [[V_COERCE0:%.*]], <vscale x 4 x i32> [[V_COERCE1:%.*]], <vscale x 4 x i32> [[V_COERCE2:%.*]], <vscale x 4 x i32> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z24test_svstnt1_vnum_u32_x4u11__SVCount_tPjl12svuint32x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CPP-CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv4i32(<vscale x 4 x i32> [[V_COERCE0:%.*]], <vscale x 4 x i32> [[V_COERCE1:%.*]], <vscale x 4 x i32> [[V_COERCE2:%.*]], <vscale x 4 x i32> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_u32_x4(svcount_t pn, uint32_t *base, int64_t vnum, svuint32x4_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_u32_x4,,)(pn, base, vnum, v);
}


// CHECK-LABEL: @test_svstnt1_vnum_u64_x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv2i64(<vscale x 2 x i64> [[V_COERCE0:%.*]], <vscale x 2 x i64> [[V_COERCE1:%.*]], <vscale x 2 x i64> [[V_COERCE2:%.*]], <vscale x 2 x i64> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z24test_svstnt1_vnum_u64_x4u11__SVCount_tPml12svuint64x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CPP-CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv2i64(<vscale x 2 x i64> [[V_COERCE0:%.*]], <vscale x 2 x i64> [[V_COERCE1:%.*]], <vscale x 2 x i64> [[V_COERCE2:%.*]], <vscale x 2 x i64> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_u64_x4(svcount_t pn, uint64_t *base, int64_t vnum, svuint64x4_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_u64_x4,,)(pn, base, vnum, v);
}


// CHECK-LABEL: @test_svstnt1_vnum_s8_x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv16i8(<vscale x 16 x i8> [[V_COERCE0:%.*]], <vscale x 16 x i8> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z23test_svstnt1_vnum_s8_x2u11__SVCount_tPal10svint8x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CPP-CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv16i8(<vscale x 16 x i8> [[V_COERCE0:%.*]], <vscale x 16 x i8> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_s8_x2(svcount_t pn, int8_t *base, int64_t vnum, svint8x2_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_s8_x2,,)(pn, base, vnum, v);
}


// CHECK-LABEL: @test_svstnt1_vnum_s16_x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv8i16(<vscale x 8 x i16> [[V_COERCE0:%.*]], <vscale x 8 x i16> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z24test_svstnt1_vnum_s16_x2u11__SVCount_tPsl11svint16x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CPP-CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv8i16(<vscale x 8 x i16> [[V_COERCE0:%.*]], <vscale x 8 x i16> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_s16_x2(svcount_t pn, int16_t *base, int64_t vnum, svint16x2_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_s16_x2,,)(pn, base, vnum, v);
}


// CHECK-LABEL: @test_svstnt1_vnum_s32_x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv4i32(<vscale x 4 x i32> [[V_COERCE0:%.*]], <vscale x 4 x i32> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z24test_svstnt1_vnum_s32_x2u11__SVCount_tPil11svint32x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CPP-CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv4i32(<vscale x 4 x i32> [[V_COERCE0:%.*]], <vscale x 4 x i32> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_s32_x2(svcount_t pn, int32_t *base, int64_t vnum, svint32x2_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_s32_x2,,)(pn, base, vnum, v);
}


// CHECK-LABEL: @test_svstnt1_vnum_s64_x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv2i64(<vscale x 2 x i64> [[V_COERCE0:%.*]], <vscale x 2 x i64> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z24test_svstnt1_vnum_s64_x2u11__SVCount_tPll11svint64x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CPP-CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv2i64(<vscale x 2 x i64> [[V_COERCE0:%.*]], <vscale x 2 x i64> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_s64_x2(svcount_t pn, int64_t *base, int64_t vnum, svint64x2_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_s64_x2,,)(pn, base, vnum, v);
}


// CHECK-LABEL: @test_svstnt1_vnum_s8_x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv16i8(<vscale x 16 x i8> [[V_COERCE0:%.*]], <vscale x 16 x i8> [[V_COERCE1:%.*]], <vscale x 16 x i8> [[V_COERCE2:%.*]], <vscale x 16 x i8> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z23test_svstnt1_vnum_s8_x4u11__SVCount_tPal10svint8x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CPP-CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv16i8(<vscale x 16 x i8> [[V_COERCE0:%.*]], <vscale x 16 x i8> [[V_COERCE1:%.*]], <vscale x 16 x i8> [[V_COERCE2:%.*]], <vscale x 16 x i8> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_s8_x4(svcount_t pn, int8_t *base, int64_t vnum, svint8x4_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_s8_x4,,)(pn, base, vnum, v);
}


// CHECK-LABEL: @test_svstnt1_vnum_s16_x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv8i16(<vscale x 8 x i16> [[V_COERCE0:%.*]], <vscale x 8 x i16> [[V_COERCE1:%.*]], <vscale x 8 x i16> [[V_COERCE2:%.*]], <vscale x 8 x i16> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z24test_svstnt1_vnum_s16_x4u11__SVCount_tPsl11svint16x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CPP-CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv8i16(<vscale x 8 x i16> [[V_COERCE0:%.*]], <vscale x 8 x i16> [[V_COERCE1:%.*]], <vscale x 8 x i16> [[V_COERCE2:%.*]], <vscale x 8 x i16> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_s16_x4(svcount_t pn, int16_t *base, int64_t vnum, svint16x4_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_s16_x4,,)(pn, base, vnum, v);
}


// CHECK-LABEL: @test_svstnt1_vnum_s32_x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv4i32(<vscale x 4 x i32> [[V_COERCE0:%.*]], <vscale x 4 x i32> [[V_COERCE1:%.*]], <vscale x 4 x i32> [[V_COERCE2:%.*]], <vscale x 4 x i32> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z24test_svstnt1_vnum_s32_x4u11__SVCount_tPil11svint32x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CPP-CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv4i32(<vscale x 4 x i32> [[V_COERCE0:%.*]], <vscale x 4 x i32> [[V_COERCE1:%.*]], <vscale x 4 x i32> [[V_COERCE2:%.*]], <vscale x 4 x i32> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_s32_x4(svcount_t pn, int32_t *base, int64_t vnum, svint32x4_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_s32_x4,,)(pn, base, vnum, v);
}


// CHECK-LABEL: @test_svstnt1_vnum_s64_x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv2i64(<vscale x 2 x i64> [[V_COERCE0:%.*]], <vscale x 2 x i64> [[V_COERCE1:%.*]], <vscale x 2 x i64> [[V_COERCE2:%.*]], <vscale x 2 x i64> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z24test_svstnt1_vnum_s64_x4u11__SVCount_tPll11svint64x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CPP-CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv2i64(<vscale x 2 x i64> [[V_COERCE0:%.*]], <vscale x 2 x i64> [[V_COERCE1:%.*]], <vscale x 2 x i64> [[V_COERCE2:%.*]], <vscale x 2 x i64> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_s64_x4(svcount_t pn, int64_t *base, int64_t vnum, svint64x4_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_s64_x4,,)(pn, base, vnum, v);
}


// CHECK-LABEL: @test_svstnt1_vnum_f16_x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[CONV:%.*]] = fptosi double [[VNUM:%.*]] to i64
// CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[CONV]]
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv8f16(<vscale x 8 x half> [[V_COERCE0:%.*]], <vscale x 8 x half> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z24test_svstnt1_vnum_f16_x2u11__SVCount_tPDhd13svfloat16x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[CONV:%.*]] = fptosi double [[VNUM:%.*]] to i64
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CPP-CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[CONV]]
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv8f16(<vscale x 8 x half> [[V_COERCE0:%.*]], <vscale x 8 x half> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_f16_x2(svcount_t pn, float16_t *base, float64_t vnum, svfloat16x2_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_f16_x2,,)(pn, base, vnum, v);
}


// CHECK-LABEL: @test_svstnt1_vnum_f32_x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[CONV:%.*]] = fptosi double [[VNUM:%.*]] to i64
// CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[CONV]]
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv4f32(<vscale x 4 x float> [[V_COERCE0:%.*]], <vscale x 4 x float> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z24test_svstnt1_vnum_f32_x2u11__SVCount_tPfd13svfloat32x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[CONV:%.*]] = fptosi double [[VNUM:%.*]] to i64
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CPP-CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[CONV]]
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv4f32(<vscale x 4 x float> [[V_COERCE0:%.*]], <vscale x 4 x float> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_f32_x2(svcount_t pn, float32_t *base, float64_t vnum, svfloat32x2_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_f32_x2,,)(pn, base, vnum, v);
}


// CHECK-LABEL: @test_svstnt1_vnum_f64_x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[CONV:%.*]] = fptosi double [[VNUM:%.*]] to i64
// CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[CONV]]
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv2f64(<vscale x 2 x double> [[V_COERCE0:%.*]], <vscale x 2 x double> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z24test_svstnt1_vnum_f64_x2u11__SVCount_tPdd13svfloat64x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[CONV:%.*]] = fptosi double [[VNUM:%.*]] to i64
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CPP-CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[CONV]]
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv2f64(<vscale x 2 x double> [[V_COERCE0:%.*]], <vscale x 2 x double> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_f64_x2(svcount_t pn, float64_t *base, float64_t vnum, svfloat64x2_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_f64_x2,,)(pn, base, vnum, v);
}

// CHECK-LABEL: @test_svstnt1_vnum_mf8_x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv16i8(<vscale x 16 x i8> [[V_COERCE0:%.*]], <vscale x 16 x i8> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z24test_svstnt1_vnum_mf8_x2u11__SVCount_tPu6__mfp8l13svmfloat8x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CPP-CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x2.nxv16i8(<vscale x 16 x i8> [[V_COERCE0:%.*]], <vscale x 16 x i8> [[V_COERCE1:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_mf8_x2(svcount_t pn, mfloat8_t *base, int64_t vnum, svmfloat8x2_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_mf8_x2,,)(pn, base, vnum, v);
}

// CHECK-LABEL: @test_svstnt1_vnum_f16_x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[CONV:%.*]] = fptosi double [[VNUM:%.*]] to i64
// CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[CONV]]
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv8f16(<vscale x 8 x half> [[V_COERCE0:%.*]], <vscale x 8 x half> [[V_COERCE1:%.*]], <vscale x 8 x half> [[V_COERCE2:%.*]], <vscale x 8 x half> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z24test_svstnt1_vnum_f16_x4u11__SVCount_tPDhd13svfloat16x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[CONV:%.*]] = fptosi double [[VNUM:%.*]] to i64
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CPP-CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[CONV]]
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv8f16(<vscale x 8 x half> [[V_COERCE0:%.*]], <vscale x 8 x half> [[V_COERCE1:%.*]], <vscale x 8 x half> [[V_COERCE2:%.*]], <vscale x 8 x half> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_f16_x4(svcount_t pn, float16_t *base, float64_t vnum, svfloat16x4_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_f16_x4,,)(pn, base, vnum, v);
}


// CHECK-LABEL: @test_svstnt1_vnum_f32_x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[CONV:%.*]] = fptosi double [[VNUM:%.*]] to i64
// CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[CONV]]
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv4f32(<vscale x 4 x float> [[V_COERCE0:%.*]], <vscale x 4 x float> [[V_COERCE1:%.*]], <vscale x 4 x float> [[V_COERCE2:%.*]], <vscale x 4 x float> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z24test_svstnt1_vnum_f32_x4u11__SVCount_tPfd13svfloat32x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[CONV:%.*]] = fptosi double [[VNUM:%.*]] to i64
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CPP-CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[CONV]]
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv4f32(<vscale x 4 x float> [[V_COERCE0:%.*]], <vscale x 4 x float> [[V_COERCE1:%.*]], <vscale x 4 x float> [[V_COERCE2:%.*]], <vscale x 4 x float> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_f32_x4(svcount_t pn, float32_t *base, float64_t vnum, svfloat32x4_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_f32_x4,,)(pn, base, vnum, v);
}


// CHECK-LABEL: @test_svstnt1_vnum_f64_x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[CONV:%.*]] = fptosi double [[VNUM:%.*]] to i64
// CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[CONV]]
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv2f64(<vscale x 2 x double> [[V_COERCE0:%.*]], <vscale x 2 x double> [[V_COERCE1:%.*]], <vscale x 2 x double> [[V_COERCE2:%.*]], <vscale x 2 x double> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z24test_svstnt1_vnum_f64_x4u11__SVCount_tPdd13svfloat64x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[CONV:%.*]] = fptosi double [[VNUM:%.*]] to i64
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CPP-CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[CONV]]
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv2f64(<vscale x 2 x double> [[V_COERCE0:%.*]], <vscale x 2 x double> [[V_COERCE1:%.*]], <vscale x 2 x double> [[V_COERCE2:%.*]], <vscale x 2 x double> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_f64_x4(svcount_t pn, float64_t *base, float64_t vnum, svfloat64x4_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_f64_x4,,)(pn, base, vnum, v);
}

// CHECK-LABEL: @test_svstnt1_vnum_mf8_x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv16i8(<vscale x 16 x i8> [[V_COERCE0:%.*]], <vscale x 16 x i8> [[V_COERCE1:%.*]], <vscale x 16 x i8> [[V_COERCE2:%.*]], <vscale x 16 x i8> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z24test_svstnt1_vnum_mf8_x4u11__SVCount_tPu6__mfp8l13svmfloat8x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call i64 @llvm.vscale.i64()
// CPP-CHECK-NEXT:    [[TMP1:%.*]] = shl nuw nsw i64 [[TMP0]], 4
// CPP-CHECK-NEXT:    [[DOTIDX:%.*]] = mul i64 [[TMP1]], [[VNUM:%.*]]
// CPP-CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[BASE:%.*]], i64 [[DOTIDX]]
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sve.stnt1.pn.x4.nxv16i8(<vscale x 16 x i8> [[V_COERCE0:%.*]], <vscale x 16 x i8> [[V_COERCE1:%.*]], <vscale x 16 x i8> [[V_COERCE2:%.*]], <vscale x 16 x i8> [[V_COERCE3:%.*]], target("aarch64.svcount") [[PN:%.*]], ptr [[TMP2]])
// CPP-CHECK-NEXT:    ret void
//
void test_svstnt1_vnum_mf8_x4(svcount_t pn, mfloat8_t *base, int64_t vnum, svmfloat8x4_t v) ATTR
{
  return SVE_ACLE_FUNC(svstnt1_vnum,_mf8_x4,,)(pn, base, vnum, v);
}