// 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 -disable-O0-optnone -Werror -Wall -emit-llvm -o - %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s
// RUN: %clang_cc1 -triple aarch64 -target-feature +sme -target-feature +sme2 -disable-O0-optnone -Werror -Wall -emit-llvm -o - -x c++ %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s -check-prefix=CPP-CHECK
// RUN: %clang_cc1 -DSVE_OVERLOADED_FORMS -triple aarch64 -target-feature +sme -target-feature +sme2 -disable-O0-optnone -Werror -Wall -emit-llvm -o - %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s
// RUN: %clang_cc1 -DSVE_OVERLOADED_FORMS -triple aarch64 -target-feature +sme -target-feature +sme2 -disable-O0-optnone -Werror -Wall -emit-llvm -o - -x c++ %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s -check-prefix=CPP-CHECK
// RUN: %clang_cc1 -triple aarch64 -target-feature +sme -target-feature +sme2 -S -disable-O0-optnone -Werror -Wall -o /dev/null %s

#include <arm_sme.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_svwrite_ver_za8_u8_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv16i8(i32 0, i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_ver_za8_u8_vg2j11svuint8x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv16i8(i32 0, i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za8_u8_vg2(uint32_t base, svuint8x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za8,_u8,_vg2,)(0, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za8_s8_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv16i8(i32 0, i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_ver_za8_s8_vg2j10svint8x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv16i8(i32 0, i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za8_s8_vg2(uint32_t base, svint8x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za8,_s8,_vg2,)(0, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za8_mf8_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv16i8(i32 0, i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z28test_svwrite_ver_za8_mf8_vg2j13svmfloat8x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv16i8(i32 0, i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za8_mf8_vg2(uint32_t base, svmfloat8x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za8,_mf8,_vg2,)(0, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za8_u8_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv16i8(i32 0, i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_hor_za8_u8_vg2j11svuint8x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv16i8(i32 0, i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za8_u8_vg2(uint32_t base, svuint8x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za8,_u8,_vg2,)(0, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za8_s8_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv16i8(i32 0, i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_hor_za8_s8_vg2j10svint8x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv16i8(i32 0, i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za8_s8_vg2(uint32_t base, svint8x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za8,_s8,_vg2,)(0, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za8_mf8_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv16i8(i32 0, i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z28test_svwrite_hor_za8_mf8_vg2j13svmfloat8x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv16i8(i32 0, i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za8_mf8_vg2(uint32_t base, svmfloat8x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za8,_mf8,_vg2,)(0, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za8_u8_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv16i8(i32 0, i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]], <vscale x 16 x i8> [[VAL_COERCE2:%.*]], <vscale x 16 x i8> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_hor_za8_u8_vg4j11svuint8x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv16i8(i32 0, i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]], <vscale x 16 x i8> [[VAL_COERCE2:%.*]], <vscale x 16 x i8> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za8_u8_vg4(uint32_t base, svuint8x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za8,_u8,_vg4,)(0, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za8_s8_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv16i8(i32 0, i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]], <vscale x 16 x i8> [[VAL_COERCE2:%.*]], <vscale x 16 x i8> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_hor_za8_s8_vg4j10svint8x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv16i8(i32 0, i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]], <vscale x 16 x i8> [[VAL_COERCE2:%.*]], <vscale x 16 x i8> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za8_s8_vg4(uint32_t base, svint8x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za8,_s8,_vg4,)(0, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za8_mf8_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv16i8(i32 0, i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]], <vscale x 16 x i8> [[VAL_COERCE2:%.*]], <vscale x 16 x i8> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z28test_svwrite_hor_za8_mf8_vg4j13svmfloat8x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv16i8(i32 0, i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]], <vscale x 16 x i8> [[VAL_COERCE2:%.*]], <vscale x 16 x i8> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za8_mf8_vg4(uint32_t base, svmfloat8x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za8,_mf8,_vg4,)(0, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za8_u8_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv16i8(i32 0, i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]], <vscale x 16 x i8> [[VAL_COERCE2:%.*]], <vscale x 16 x i8> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_ver_za8_u8_vg4j11svuint8x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv16i8(i32 0, i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]], <vscale x 16 x i8> [[VAL_COERCE2:%.*]], <vscale x 16 x i8> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za8_u8_vg4(uint32_t base, svuint8x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za8,_u8,_vg4,)(0, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za8_s8_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv16i8(i32 0, i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]], <vscale x 16 x i8> [[VAL_COERCE2:%.*]], <vscale x 16 x i8> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_ver_za8_s8_vg4j10svint8x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv16i8(i32 0, i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]], <vscale x 16 x i8> [[VAL_COERCE2:%.*]], <vscale x 16 x i8> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za8_s8_vg4(uint32_t base, svint8x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za8,_s8,_vg4,)(0, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za8_mf8_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv16i8(i32 0, i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]], <vscale x 16 x i8> [[VAL_COERCE2:%.*]], <vscale x 16 x i8> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z28test_svwrite_ver_za8_mf8_vg4j13svmfloat8x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv16i8(i32 0, i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]], <vscale x 16 x i8> [[VAL_COERCE2:%.*]], <vscale x 16 x i8> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za8_mf8_vg4(uint32_t base, svmfloat8x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za8,_mf8,_vg4,)(0, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za16_u16_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv8i16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x i16> [[VAL_COERCE0:%.*]], <vscale x 8 x i16> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_hor_za16_u16_vg2j12svuint16x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv8i16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x i16> [[VAL_COERCE0:%.*]], <vscale x 8 x i16> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za16_u16_vg2(uint32_t base, svuint16x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za16,_u16,_vg2,)(1, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za16_bf16_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv8bf16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE0:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z30test_svwrite_hor_za16_bf16_vg2j14svbfloat16x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv8bf16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE0:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za16_bf16_vg2(uint32_t base, svbfloat16x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za16,_bf16,_vg2,)(1, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za16_f16_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv8f16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x half> [[VAL_COERCE0:%.*]], <vscale x 8 x half> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_hor_za16_f16_vg2j13svfloat16x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv8f16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x half> [[VAL_COERCE0:%.*]], <vscale x 8 x half> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za16_f16_vg2(uint32_t base, svfloat16x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za16,_f16,_vg2,)(1, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za16_s16_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv8i16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x i16> [[VAL_COERCE0:%.*]], <vscale x 8 x i16> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_hor_za16_s16_vg2j11svint16x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv8i16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x i16> [[VAL_COERCE0:%.*]], <vscale x 8 x i16> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za16_s16_vg2(uint32_t base, svint16x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za16,_s16,_vg2,)(1, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za16_u16_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv8i16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x i16> [[VAL_COERCE0:%.*]], <vscale x 8 x i16> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_ver_za16_u16_vg2j12svuint16x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv8i16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x i16> [[VAL_COERCE0:%.*]], <vscale x 8 x i16> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za16_u16_vg2(uint32_t base, svuint16x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za16,_u16,_vg2,)(1, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za16_bf16_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv8bf16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE0:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z30test_svwrite_ver_za16_bf16_vg2j14svbfloat16x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv8bf16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE0:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za16_bf16_vg2(uint32_t base, svbfloat16x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za16,_bf16,_vg2,)(1, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za16_f16_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv8f16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x half> [[VAL_COERCE0:%.*]], <vscale x 8 x half> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_ver_za16_f16_vg2j13svfloat16x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv8f16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x half> [[VAL_COERCE0:%.*]], <vscale x 8 x half> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za16_f16_vg2(uint32_t base, svfloat16x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za16,_f16,_vg2,)(1, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za16_s16_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv8i16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x i16> [[VAL_COERCE0:%.*]], <vscale x 8 x i16> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_ver_za16_s16_vg2j11svint16x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv8i16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x i16> [[VAL_COERCE0:%.*]], <vscale x 8 x i16> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za16_s16_vg2(uint32_t base, svint16x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za16,_s16,_vg2,)(1, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za16_u16_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv8i16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x i16> [[VAL_COERCE0:%.*]], <vscale x 8 x i16> [[VAL_COERCE1:%.*]], <vscale x 8 x i16> [[VAL_COERCE2:%.*]], <vscale x 8 x i16> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_hor_za16_u16_vg4j12svuint16x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv8i16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x i16> [[VAL_COERCE0:%.*]], <vscale x 8 x i16> [[VAL_COERCE1:%.*]], <vscale x 8 x i16> [[VAL_COERCE2:%.*]], <vscale x 8 x i16> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za16_u16_vg4(uint32_t base, svuint16x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za16,_u16,_vg4,)(1, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za16_bf16_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv8bf16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE0:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE1:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE2:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z30test_svwrite_hor_za16_bf16_vg4j14svbfloat16x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv8bf16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE0:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE1:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE2:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za16_bf16_vg4(uint32_t base, svbfloat16x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za16,_bf16,_vg4,)(1, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za16_f16_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv8f16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x half> [[VAL_COERCE0:%.*]], <vscale x 8 x half> [[VAL_COERCE1:%.*]], <vscale x 8 x half> [[VAL_COERCE2:%.*]], <vscale x 8 x half> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_hor_za16_f16_vg4j13svfloat16x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv8f16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x half> [[VAL_COERCE0:%.*]], <vscale x 8 x half> [[VAL_COERCE1:%.*]], <vscale x 8 x half> [[VAL_COERCE2:%.*]], <vscale x 8 x half> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za16_f16_vg4(uint32_t base, svfloat16x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za16,_f16,_vg4,)(1, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za16_s16_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv8i16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x i16> [[VAL_COERCE0:%.*]], <vscale x 8 x i16> [[VAL_COERCE1:%.*]], <vscale x 8 x i16> [[VAL_COERCE2:%.*]], <vscale x 8 x i16> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_hor_za16_s16_vg4j11svint16x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv8i16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x i16> [[VAL_COERCE0:%.*]], <vscale x 8 x i16> [[VAL_COERCE1:%.*]], <vscale x 8 x i16> [[VAL_COERCE2:%.*]], <vscale x 8 x i16> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za16_s16_vg4(uint32_t base, svint16x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za16,_s16,_vg4,)(1, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za16_u16_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv8i16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x i16> [[VAL_COERCE0:%.*]], <vscale x 8 x i16> [[VAL_COERCE1:%.*]], <vscale x 8 x i16> [[VAL_COERCE2:%.*]], <vscale x 8 x i16> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_ver_za16_u16_vg4j12svuint16x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv8i16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x i16> [[VAL_COERCE0:%.*]], <vscale x 8 x i16> [[VAL_COERCE1:%.*]], <vscale x 8 x i16> [[VAL_COERCE2:%.*]], <vscale x 8 x i16> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za16_u16_vg4(uint32_t base, svuint16x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za16,_u16,_vg4,)(1, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za16_bf16_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv8bf16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE0:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE1:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE2:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z30test_svwrite_ver_za16_bf16_vg4j14svbfloat16x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv8bf16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE0:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE1:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE2:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za16_bf16_vg4(uint32_t base, svbfloat16x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za16,_bf16,_vg4,)(1, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za16_f16_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv8f16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x half> [[VAL_COERCE0:%.*]], <vscale x 8 x half> [[VAL_COERCE1:%.*]], <vscale x 8 x half> [[VAL_COERCE2:%.*]], <vscale x 8 x half> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_ver_za16_f16_vg4j13svfloat16x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv8f16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x half> [[VAL_COERCE0:%.*]], <vscale x 8 x half> [[VAL_COERCE1:%.*]], <vscale x 8 x half> [[VAL_COERCE2:%.*]], <vscale x 8 x half> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za16_f16_vg4(uint32_t base, svfloat16x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za16,_f16,_vg4,)(1, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za16_s16_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv8i16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x i16> [[VAL_COERCE0:%.*]], <vscale x 8 x i16> [[VAL_COERCE1:%.*]], <vscale x 8 x i16> [[VAL_COERCE2:%.*]], <vscale x 8 x i16> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_ver_za16_s16_vg4j11svint16x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv8i16(i32 1, i32 [[BASE:%.*]], <vscale x 8 x i16> [[VAL_COERCE0:%.*]], <vscale x 8 x i16> [[VAL_COERCE1:%.*]], <vscale x 8 x i16> [[VAL_COERCE2:%.*]], <vscale x 8 x i16> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za16_s16_vg4(uint32_t base, svint16x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za16,_s16,_vg4,)(1, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za32_u32_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv4i32(i32 3, i32 [[BASE:%.*]], <vscale x 4 x i32> [[VAL_COERCE0:%.*]], <vscale x 4 x i32> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_hor_za32_u32_vg2j12svuint32x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv4i32(i32 3, i32 [[BASE:%.*]], <vscale x 4 x i32> [[VAL_COERCE0:%.*]], <vscale x 4 x i32> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za32_u32_vg2(uint32_t base, svuint32x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za32,_u32,_vg2,)(3, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za32_f32_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv4f32(i32 3, i32 [[BASE:%.*]], <vscale x 4 x float> [[VAL_COERCE0:%.*]], <vscale x 4 x float> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_hor_za32_f32_vg2j13svfloat32x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv4f32(i32 3, i32 [[BASE:%.*]], <vscale x 4 x float> [[VAL_COERCE0:%.*]], <vscale x 4 x float> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za32_f32_vg2(uint32_t base, svfloat32x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za32,_f32,_vg2,)(3, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za32_s32_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv4i32(i32 3, i32 [[BASE:%.*]], <vscale x 4 x i32> [[VAL_COERCE0:%.*]], <vscale x 4 x i32> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_hor_za32_s32_vg2j11svint32x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv4i32(i32 3, i32 [[BASE:%.*]], <vscale x 4 x i32> [[VAL_COERCE0:%.*]], <vscale x 4 x i32> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za32_s32_vg2(uint32_t base, svint32x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za32,_s32,_vg2,)(3, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za32_u32_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv4i32(i32 3, i32 [[BASE:%.*]], <vscale x 4 x i32> [[VAL_COERCE0:%.*]], <vscale x 4 x i32> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_ver_za32_u32_vg2j12svuint32x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv4i32(i32 3, i32 [[BASE:%.*]], <vscale x 4 x i32> [[VAL_COERCE0:%.*]], <vscale x 4 x i32> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za32_u32_vg2(uint32_t base, svuint32x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za32,_u32,_vg2,)(3, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za32_f32_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv4f32(i32 3, i32 [[BASE:%.*]], <vscale x 4 x float> [[VAL_COERCE0:%.*]], <vscale x 4 x float> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_ver_za32_f32_vg2j13svfloat32x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv4f32(i32 3, i32 [[BASE:%.*]], <vscale x 4 x float> [[VAL_COERCE0:%.*]], <vscale x 4 x float> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za32_f32_vg2(uint32_t base, svfloat32x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za32,_f32,_vg2,)(3, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za32_s32_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv4i32(i32 3, i32 [[BASE:%.*]], <vscale x 4 x i32> [[VAL_COERCE0:%.*]], <vscale x 4 x i32> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_ver_za32_s32_vg2j11svint32x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv4i32(i32 3, i32 [[BASE:%.*]], <vscale x 4 x i32> [[VAL_COERCE0:%.*]], <vscale x 4 x i32> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za32_s32_vg2(uint32_t base, svint32x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za32,_s32,_vg2,)(3, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za32_u32_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv4i32(i32 3, i32 [[BASE:%.*]], <vscale x 4 x i32> [[VAL_COERCE0:%.*]], <vscale x 4 x i32> [[VAL_COERCE1:%.*]], <vscale x 4 x i32> [[VAL_COERCE2:%.*]], <vscale x 4 x i32> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_hor_za32_u32_vg4j12svuint32x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv4i32(i32 3, i32 [[BASE:%.*]], <vscale x 4 x i32> [[VAL_COERCE0:%.*]], <vscale x 4 x i32> [[VAL_COERCE1:%.*]], <vscale x 4 x i32> [[VAL_COERCE2:%.*]], <vscale x 4 x i32> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za32_u32_vg4(uint32_t base, svuint32x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za32,_u32,_vg4,)(3, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za32_f32_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv4f32(i32 3, i32 [[BASE:%.*]], <vscale x 4 x float> [[VAL_COERCE0:%.*]], <vscale x 4 x float> [[VAL_COERCE1:%.*]], <vscale x 4 x float> [[VAL_COERCE2:%.*]], <vscale x 4 x float> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_hor_za32_f32_vg4j13svfloat32x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv4f32(i32 3, i32 [[BASE:%.*]], <vscale x 4 x float> [[VAL_COERCE0:%.*]], <vscale x 4 x float> [[VAL_COERCE1:%.*]], <vscale x 4 x float> [[VAL_COERCE2:%.*]], <vscale x 4 x float> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za32_f32_vg4(uint32_t base, svfloat32x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za32,_f32,_vg4,)(3, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za32_s32_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv4i32(i32 3, i32 [[BASE:%.*]], <vscale x 4 x i32> [[VAL_COERCE0:%.*]], <vscale x 4 x i32> [[VAL_COERCE1:%.*]], <vscale x 4 x i32> [[VAL_COERCE2:%.*]], <vscale x 4 x i32> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_hor_za32_s32_vg4j11svint32x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv4i32(i32 3, i32 [[BASE:%.*]], <vscale x 4 x i32> [[VAL_COERCE0:%.*]], <vscale x 4 x i32> [[VAL_COERCE1:%.*]], <vscale x 4 x i32> [[VAL_COERCE2:%.*]], <vscale x 4 x i32> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za32_s32_vg4(uint32_t base, svint32x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za32,_s32,_vg4,)(3, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za32_u32_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv4i32(i32 3, i32 [[BASE:%.*]], <vscale x 4 x i32> [[VAL_COERCE0:%.*]], <vscale x 4 x i32> [[VAL_COERCE1:%.*]], <vscale x 4 x i32> [[VAL_COERCE2:%.*]], <vscale x 4 x i32> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_ver_za32_u32_vg4j12svuint32x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv4i32(i32 3, i32 [[BASE:%.*]], <vscale x 4 x i32> [[VAL_COERCE0:%.*]], <vscale x 4 x i32> [[VAL_COERCE1:%.*]], <vscale x 4 x i32> [[VAL_COERCE2:%.*]], <vscale x 4 x i32> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za32_u32_vg4(uint32_t base, svuint32x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za32,_u32,_vg4,)(3, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za32_f32_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv4f32(i32 3, i32 [[BASE:%.*]], <vscale x 4 x float> [[VAL_COERCE0:%.*]], <vscale x 4 x float> [[VAL_COERCE1:%.*]], <vscale x 4 x float> [[VAL_COERCE2:%.*]], <vscale x 4 x float> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_ver_za32_f32_vg4j13svfloat32x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv4f32(i32 3, i32 [[BASE:%.*]], <vscale x 4 x float> [[VAL_COERCE0:%.*]], <vscale x 4 x float> [[VAL_COERCE1:%.*]], <vscale x 4 x float> [[VAL_COERCE2:%.*]], <vscale x 4 x float> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za32_f32_vg4(uint32_t base, svfloat32x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za32,_f32,_vg4,)(3, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za32_s32_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv4i32(i32 3, i32 [[BASE:%.*]], <vscale x 4 x i32> [[VAL_COERCE0:%.*]], <vscale x 4 x i32> [[VAL_COERCE1:%.*]], <vscale x 4 x i32> [[VAL_COERCE2:%.*]], <vscale x 4 x i32> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_ver_za32_s32_vg4j11svint32x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv4i32(i32 3, i32 [[BASE:%.*]], <vscale x 4 x i32> [[VAL_COERCE0:%.*]], <vscale x 4 x i32> [[VAL_COERCE1:%.*]], <vscale x 4 x i32> [[VAL_COERCE2:%.*]], <vscale x 4 x i32> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za32_s32_vg4(uint32_t base, svint32x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za32,_s32,_vg4,)(3, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za64_u64_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv2i64(i32 7, i32 [[BASE:%.*]], <vscale x 2 x i64> [[VAL_COERCE0:%.*]], <vscale x 2 x i64> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_hor_za64_u64_vg2j12svuint64x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv2i64(i32 7, i32 [[BASE:%.*]], <vscale x 2 x i64> [[VAL_COERCE0:%.*]], <vscale x 2 x i64> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za64_u64_vg2(uint32_t base, svuint64x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za64,_u64,_vg2,)(7, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za64_f64_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv2f64(i32 7, i32 [[BASE:%.*]], <vscale x 2 x double> [[VAL_COERCE0:%.*]], <vscale x 2 x double> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_hor_za64_f64_vg2j13svfloat64x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv2f64(i32 7, i32 [[BASE:%.*]], <vscale x 2 x double> [[VAL_COERCE0:%.*]], <vscale x 2 x double> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za64_f64_vg2(uint32_t base, svfloat64x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za64,_f64,_vg2,)(7, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za64_s64_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv2i64(i32 7, i32 [[BASE:%.*]], <vscale x 2 x i64> [[VAL_COERCE0:%.*]], <vscale x 2 x i64> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_hor_za64_s64_vg2j11svint64x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg2.nxv2i64(i32 7, i32 [[BASE:%.*]], <vscale x 2 x i64> [[VAL_COERCE0:%.*]], <vscale x 2 x i64> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za64_s64_vg2(uint32_t base, svint64x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za64,_s64,_vg2,)(7, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za64_u64_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv2i64(i32 7, i32 [[BASE:%.*]], <vscale x 2 x i64> [[VAL_COERCE0:%.*]], <vscale x 2 x i64> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_ver_za64_u64_vg2j12svuint64x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv2i64(i32 7, i32 [[BASE:%.*]], <vscale x 2 x i64> [[VAL_COERCE0:%.*]], <vscale x 2 x i64> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za64_u64_vg2(uint32_t base, svuint64x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za64,_u64,_vg2,)(7, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za64_f64_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv2f64(i32 7, i32 [[BASE:%.*]], <vscale x 2 x double> [[VAL_COERCE0:%.*]], <vscale x 2 x double> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_ver_za64_f64_vg2j13svfloat64x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv2f64(i32 7, i32 [[BASE:%.*]], <vscale x 2 x double> [[VAL_COERCE0:%.*]], <vscale x 2 x double> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za64_f64_vg2(uint32_t base, svfloat64x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za64,_f64,_vg2,)(7, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za64_s64_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv2i64(i32 7, i32 [[BASE:%.*]], <vscale x 2 x i64> [[VAL_COERCE0:%.*]], <vscale x 2 x i64> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_ver_za64_s64_vg2j11svint64x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg2.nxv2i64(i32 7, i32 [[BASE:%.*]], <vscale x 2 x i64> [[VAL_COERCE0:%.*]], <vscale x 2 x i64> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za64_s64_vg2(uint32_t base, svint64x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za64,_s64,_vg2,)(7, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za64_u64_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv2i64(i32 7, i32 [[BASE:%.*]], <vscale x 2 x i64> [[VAL_COERCE0:%.*]], <vscale x 2 x i64> [[VAL_COERCE1:%.*]], <vscale x 2 x i64> [[VAL_COERCE2:%.*]], <vscale x 2 x i64> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_hor_za64_u64_vg4j12svuint64x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv2i64(i32 7, i32 [[BASE:%.*]], <vscale x 2 x i64> [[VAL_COERCE0:%.*]], <vscale x 2 x i64> [[VAL_COERCE1:%.*]], <vscale x 2 x i64> [[VAL_COERCE2:%.*]], <vscale x 2 x i64> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za64_u64_vg4(uint32_t base, svuint64x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za64,_u64,_vg4,)(7, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za64_u64_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv2i64(i32 7, i32 [[BASE:%.*]], <vscale x 2 x i64> [[VAL_COERCE0:%.*]], <vscale x 2 x i64> [[VAL_COERCE1:%.*]], <vscale x 2 x i64> [[VAL_COERCE2:%.*]], <vscale x 2 x i64> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_ver_za64_u64_vg4j12svuint64x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv2i64(i32 7, i32 [[BASE:%.*]], <vscale x 2 x i64> [[VAL_COERCE0:%.*]], <vscale x 2 x i64> [[VAL_COERCE1:%.*]], <vscale x 2 x i64> [[VAL_COERCE2:%.*]], <vscale x 2 x i64> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za64_u64_vg4(uint32_t base, svuint64x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za64,_u64,_vg4,)(7, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za64_f64_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv2f64(i32 7, i32 [[BASE:%.*]], <vscale x 2 x double> [[VAL_COERCE0:%.*]], <vscale x 2 x double> [[VAL_COERCE1:%.*]], <vscale x 2 x double> [[VAL_COERCE2:%.*]], <vscale x 2 x double> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_hor_za64_f64_vg4j13svfloat64x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv2f64(i32 7, i32 [[BASE:%.*]], <vscale x 2 x double> [[VAL_COERCE0:%.*]], <vscale x 2 x double> [[VAL_COERCE1:%.*]], <vscale x 2 x double> [[VAL_COERCE2:%.*]], <vscale x 2 x double> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za64_f64_vg4(uint32_t base, svfloat64x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za64,_f64,_vg4,)(7, base, val);
}

// CHECK-LABEL: @test_svwrite_hor_za64_s64_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv2i64(i32 7, i32 [[BASE:%.*]], <vscale x 2 x i64> [[VAL_COERCE0:%.*]], <vscale x 2 x i64> [[VAL_COERCE1:%.*]], <vscale x 2 x i64> [[VAL_COERCE2:%.*]], <vscale x 2 x i64> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_hor_za64_s64_vg4j11svint64x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.hor.vg4.nxv2i64(i32 7, i32 [[BASE:%.*]], <vscale x 2 x i64> [[VAL_COERCE0:%.*]], <vscale x 2 x i64> [[VAL_COERCE1:%.*]], <vscale x 2 x i64> [[VAL_COERCE2:%.*]], <vscale x 2 x i64> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_hor_za64_s64_vg4(uint32_t base, svint64x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_hor_za64,_s64,_vg4,)(7, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za64_f64_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv2f64(i32 7, i32 [[BASE:%.*]], <vscale x 2 x double> [[VAL_COERCE0:%.*]], <vscale x 2 x double> [[VAL_COERCE1:%.*]], <vscale x 2 x double> [[VAL_COERCE2:%.*]], <vscale x 2 x double> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_ver_za64_f64_vg4j13svfloat64x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv2f64(i32 7, i32 [[BASE:%.*]], <vscale x 2 x double> [[VAL_COERCE0:%.*]], <vscale x 2 x double> [[VAL_COERCE1:%.*]], <vscale x 2 x double> [[VAL_COERCE2:%.*]], <vscale x 2 x double> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za64_f64_vg4(uint32_t base, svfloat64x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za64,_f64,_vg4,)(7, base, val);
}

// CHECK-LABEL: @test_svwrite_ver_za64_s64_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv2i64(i32 7, i32 [[BASE:%.*]], <vscale x 2 x i64> [[VAL_COERCE0:%.*]], <vscale x 2 x i64> [[VAL_COERCE1:%.*]], <vscale x 2 x i64> [[VAL_COERCE2:%.*]], <vscale x 2 x i64> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z29test_svwrite_ver_za64_s64_vg4j11svint64x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.ver.vg4.nxv2i64(i32 7, i32 [[BASE:%.*]], <vscale x 2 x i64> [[VAL_COERCE0:%.*]], <vscale x 2 x i64> [[VAL_COERCE1:%.*]], <vscale x 2 x i64> [[VAL_COERCE2:%.*]], <vscale x 2 x i64> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_ver_za64_s64_vg4(uint32_t base, svint64x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_ver_za64,_s64,_vg4,)(7, base, val);
}

// CHECK-LABEL: @test_svwrite_za8_s8_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv16i8(i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z25test_svwrite_za8_s8_vg1x2j10svint8x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv16i8(i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za8_s8_vg1x2(uint32_t base, svint8x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za8,_s8,_vg1x2,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za8_u8_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv16i8(i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z25test_svwrite_za8_u8_vg1x2j11svuint8x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv16i8(i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za8_u8_vg1x2(uint32_t base, svuint8x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za8,_u8,_vg1x2,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za8_mf8_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv16i8(i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z26test_svwrite_za8_mf8_vg1x2j13svmfloat8x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv16i8(i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za8_mf8_vg1x2(uint32_t base, svmfloat8x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za8,_mf8,_vg1x2,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za16_s16_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv8i16(i32 [[BASE:%.*]], <vscale x 8 x i16> [[VAL_COERCE0:%.*]], <vscale x 8 x i16> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_za16_s16_vg1x2j11svint16x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv8i16(i32 [[BASE:%.*]], <vscale x 8 x i16> [[VAL_COERCE0:%.*]], <vscale x 8 x i16> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za16_s16_vg1x2(uint32_t base, svint16x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za16,_s16,_vg1x2,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za16_u16_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv8i16(i32 [[BASE:%.*]], <vscale x 8 x i16> [[VAL_COERCE0:%.*]], <vscale x 8 x i16> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_za16_u16_vg1x2j12svuint16x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv8i16(i32 [[BASE:%.*]], <vscale x 8 x i16> [[VAL_COERCE0:%.*]], <vscale x 8 x i16> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za16_u16_vg1x2(uint32_t base, svuint16x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za16,_u16,_vg1x2,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za16_bf16_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv8bf16(i32 [[BASE:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE0:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z28test_svwrite_za16_bf16_vg1x2j14svbfloat16x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv8bf16(i32 [[BASE:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE0:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za16_bf16_vg1x2(uint32_t base, svbfloat16x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za16,_bf16,_vg1x2,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za16_f16_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv8f16(i32 [[BASE:%.*]], <vscale x 8 x half> [[VAL_COERCE0:%.*]], <vscale x 8 x half> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_za16_f16_vg1x2j13svfloat16x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv8f16(i32 [[BASE:%.*]], <vscale x 8 x half> [[VAL_COERCE0:%.*]], <vscale x 8 x half> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za16_f16_vg1x2(uint32_t base, svfloat16x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za16,_f16,_vg1x2,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za32_s32_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv4i32(i32 [[BASE:%.*]], <vscale x 4 x i32> [[VAL_COERCE0:%.*]], <vscale x 4 x i32> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_za32_s32_vg1x2j11svint32x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv4i32(i32 [[BASE:%.*]], <vscale x 4 x i32> [[VAL_COERCE0:%.*]], <vscale x 4 x i32> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za32_s32_vg1x2(uint32_t base, svint32x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za32,_s32,_vg1x2,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za32_u32_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv4i32(i32 [[BASE:%.*]], <vscale x 4 x i32> [[VAL_COERCE0:%.*]], <vscale x 4 x i32> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_za32_u32_vg1x2j12svuint32x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv4i32(i32 [[BASE:%.*]], <vscale x 4 x i32> [[VAL_COERCE0:%.*]], <vscale x 4 x i32> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za32_u32_vg1x2(uint32_t base, svuint32x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za32,_u32,_vg1x2,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za32_f32_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv4f32(i32 [[BASE:%.*]], <vscale x 4 x float> [[VAL_COERCE0:%.*]], <vscale x 4 x float> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_za32_f32_vg1x2j13svfloat32x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv4f32(i32 [[BASE:%.*]], <vscale x 4 x float> [[VAL_COERCE0:%.*]], <vscale x 4 x float> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za32_f32_vg1x2(uint32_t base, svfloat32x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za32,_f32,_vg1x2,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za64_u64_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv2i64(i32 [[BASE:%.*]], <vscale x 2 x i64> [[VAL_COERCE0:%.*]], <vscale x 2 x i64> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_za64_u64_vg1x2j12svuint64x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv2i64(i32 [[BASE:%.*]], <vscale x 2 x i64> [[VAL_COERCE0:%.*]], <vscale x 2 x i64> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za64_u64_vg1x2(uint32_t base, svuint64x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za64,_u64,_vg1x2,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za64_f64_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv2f64(i32 [[BASE:%.*]], <vscale x 2 x double> [[VAL_COERCE0:%.*]], <vscale x 2 x double> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_za64_f64_vg1x2j13svfloat64x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv2f64(i32 [[BASE:%.*]], <vscale x 2 x double> [[VAL_COERCE0:%.*]], <vscale x 2 x double> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za64_f64_vg1x2(uint32_t base, svfloat64x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za64,_f64,_vg1x2,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za64_s64_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv2i64(i32 [[BASE:%.*]], <vscale x 2 x i64> [[VAL_COERCE0:%.*]], <vscale x 2 x i64> [[VAL_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_za64_s64_vg1x2j11svint64x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x2.nxv2i64(i32 [[BASE:%.*]], <vscale x 2 x i64> [[VAL_COERCE0:%.*]], <vscale x 2 x i64> [[VAL_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za64_s64_vg1x2(uint32_t base, svint64x2_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za64,_s64,_vg1x2,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za8_mf8_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv16i8(i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]], <vscale x 16 x i8> [[VAL_COERCE2:%.*]], <vscale x 16 x i8> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z26test_svwrite_za8_mf8_vg1x4j13svmfloat8x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv16i8(i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]], <vscale x 16 x i8> [[VAL_COERCE2:%.*]], <vscale x 16 x i8> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za8_mf8_vg1x4(uint32_t base, svmfloat8x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za8,_mf8,_vg1x4,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za8_s8_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv16i8(i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]], <vscale x 16 x i8> [[VAL_COERCE2:%.*]], <vscale x 16 x i8> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z25test_svwrite_za8_s8_vg1x4j10svint8x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv16i8(i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]], <vscale x 16 x i8> [[VAL_COERCE2:%.*]], <vscale x 16 x i8> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za8_s8_vg1x4(uint32_t base, svint8x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za8,_s8,_vg1x4,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za8_u8_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv16i8(i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]], <vscale x 16 x i8> [[VAL_COERCE2:%.*]], <vscale x 16 x i8> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z25test_svwrite_za8_u8_vg1x4j11svuint8x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv16i8(i32 [[BASE:%.*]], <vscale x 16 x i8> [[VAL_COERCE0:%.*]], <vscale x 16 x i8> [[VAL_COERCE1:%.*]], <vscale x 16 x i8> [[VAL_COERCE2:%.*]], <vscale x 16 x i8> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za8_u8_vg1x4(uint32_t base, svuint8x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za8,_u8,_vg1x4,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za16_s16_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv8i16(i32 [[BASE:%.*]], <vscale x 8 x i16> [[VAL_COERCE0:%.*]], <vscale x 8 x i16> [[VAL_COERCE1:%.*]], <vscale x 8 x i16> [[VAL_COERCE2:%.*]], <vscale x 8 x i16> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_za16_s16_vg1x4j11svint16x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv8i16(i32 [[BASE:%.*]], <vscale x 8 x i16> [[VAL_COERCE0:%.*]], <vscale x 8 x i16> [[VAL_COERCE1:%.*]], <vscale x 8 x i16> [[VAL_COERCE2:%.*]], <vscale x 8 x i16> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za16_s16_vg1x4(uint32_t base, svint16x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za16,_s16,_vg1x4,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za16_u16_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv8i16(i32 [[BASE:%.*]], <vscale x 8 x i16> [[VAL_COERCE0:%.*]], <vscale x 8 x i16> [[VAL_COERCE1:%.*]], <vscale x 8 x i16> [[VAL_COERCE2:%.*]], <vscale x 8 x i16> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_za16_u16_vg1x4j12svuint16x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv8i16(i32 [[BASE:%.*]], <vscale x 8 x i16> [[VAL_COERCE0:%.*]], <vscale x 8 x i16> [[VAL_COERCE1:%.*]], <vscale x 8 x i16> [[VAL_COERCE2:%.*]], <vscale x 8 x i16> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za16_u16_vg1x4(uint32_t base, svuint16x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za16,_u16,_vg1x4,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za16_bf16_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv8bf16(i32 [[BASE:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE0:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE1:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE2:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z28test_svwrite_za16_bf16_vg1x4j14svbfloat16x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv8bf16(i32 [[BASE:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE0:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE1:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE2:%.*]], <vscale x 8 x bfloat> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za16_bf16_vg1x4(uint32_t base, svbfloat16x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za16,_bf16,_vg1x4,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za16_f16_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv8f16(i32 [[BASE:%.*]], <vscale x 8 x half> [[VAL_COERCE0:%.*]], <vscale x 8 x half> [[VAL_COERCE1:%.*]], <vscale x 8 x half> [[VAL_COERCE2:%.*]], <vscale x 8 x half> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_za16_f16_vg1x4j13svfloat16x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv8f16(i32 [[BASE:%.*]], <vscale x 8 x half> [[VAL_COERCE0:%.*]], <vscale x 8 x half> [[VAL_COERCE1:%.*]], <vscale x 8 x half> [[VAL_COERCE2:%.*]], <vscale x 8 x half> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za16_f16_vg1x4(uint32_t base, svfloat16x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za16,_f16,_vg1x4,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za32_s32_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv4i32(i32 [[BASE:%.*]], <vscale x 4 x i32> [[VAL_COERCE0:%.*]], <vscale x 4 x i32> [[VAL_COERCE1:%.*]], <vscale x 4 x i32> [[VAL_COERCE2:%.*]], <vscale x 4 x i32> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_za32_s32_vg1x4j11svint32x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv4i32(i32 [[BASE:%.*]], <vscale x 4 x i32> [[VAL_COERCE0:%.*]], <vscale x 4 x i32> [[VAL_COERCE1:%.*]], <vscale x 4 x i32> [[VAL_COERCE2:%.*]], <vscale x 4 x i32> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za32_s32_vg1x4(uint32_t base, svint32x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za32,_s32,_vg1x4,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za32_u32_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv4i32(i32 [[BASE:%.*]], <vscale x 4 x i32> [[VAL_COERCE0:%.*]], <vscale x 4 x i32> [[VAL_COERCE1:%.*]], <vscale x 4 x i32> [[VAL_COERCE2:%.*]], <vscale x 4 x i32> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_za32_u32_vg1x4j12svuint32x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv4i32(i32 [[BASE:%.*]], <vscale x 4 x i32> [[VAL_COERCE0:%.*]], <vscale x 4 x i32> [[VAL_COERCE1:%.*]], <vscale x 4 x i32> [[VAL_COERCE2:%.*]], <vscale x 4 x i32> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za32_u32_vg1x4(uint32_t base, svuint32x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za32,_u32,_vg1x4,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za32_f32_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv4f32(i32 [[BASE:%.*]], <vscale x 4 x float> [[VAL_COERCE0:%.*]], <vscale x 4 x float> [[VAL_COERCE1:%.*]], <vscale x 4 x float> [[VAL_COERCE2:%.*]], <vscale x 4 x float> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_za32_f32_vg1x4j13svfloat32x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv4f32(i32 [[BASE:%.*]], <vscale x 4 x float> [[VAL_COERCE0:%.*]], <vscale x 4 x float> [[VAL_COERCE1:%.*]], <vscale x 4 x float> [[VAL_COERCE2:%.*]], <vscale x 4 x float> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za32_f32_vg1x4(uint32_t base, svfloat32x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za32,_f32,_vg1x4,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za64_u64_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv2i64(i32 [[BASE:%.*]], <vscale x 2 x i64> [[VAL_COERCE0:%.*]], <vscale x 2 x i64> [[VAL_COERCE1:%.*]], <vscale x 2 x i64> [[VAL_COERCE2:%.*]], <vscale x 2 x i64> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_za64_u64_vg1x4j12svuint64x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv2i64(i32 [[BASE:%.*]], <vscale x 2 x i64> [[VAL_COERCE0:%.*]], <vscale x 2 x i64> [[VAL_COERCE1:%.*]], <vscale x 2 x i64> [[VAL_COERCE2:%.*]], <vscale x 2 x i64> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za64_u64_vg1x4(uint32_t base, svuint64x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za64,_u64,_vg1x4,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za64_f64_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv2f64(i32 [[BASE:%.*]], <vscale x 2 x double> [[VAL_COERCE0:%.*]], <vscale x 2 x double> [[VAL_COERCE1:%.*]], <vscale x 2 x double> [[VAL_COERCE2:%.*]], <vscale x 2 x double> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_za64_f64_vg1x4j13svfloat64x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv2f64(i32 [[BASE:%.*]], <vscale x 2 x double> [[VAL_COERCE0:%.*]], <vscale x 2 x double> [[VAL_COERCE1:%.*]], <vscale x 2 x double> [[VAL_COERCE2:%.*]], <vscale x 2 x double> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za64_f64_vg1x4(uint32_t base, svfloat64x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za64,_f64,_vg1x4,)(base, val);
}

// CHECK-LABEL: @test_svwrite_za64_s64_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv2i64(i32 [[BASE:%.*]], <vscale x 2 x i64> [[VAL_COERCE0:%.*]], <vscale x 2 x i64> [[VAL_COERCE1:%.*]], <vscale x 2 x i64> [[VAL_COERCE2:%.*]], <vscale x 2 x i64> [[VAL_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svwrite_za64_s64_vg1x4j11svint64x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.write.vg1x4.nxv2i64(i32 [[BASE:%.*]], <vscale x 2 x i64> [[VAL_COERCE0:%.*]], <vscale x 2 x i64> [[VAL_COERCE1:%.*]], <vscale x 2 x i64> [[VAL_COERCE2:%.*]], <vscale x 2 x i64> [[VAL_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svwrite_za64_s64_vg1x4(uint32_t base, svint64x4_t val) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svwrite_za64,_s64,_vg1x4,)(base, val);
}