// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py

// REQUIRES: aarch64-registered-target

// RUN: %clang_cc1 -fclang-abi-compat=latest -triple aarch64 -target-feature +sme -target-feature +sme2 -target-feature +sme-i16i64 -target-feature +sme-f64f64 -disable-O0-optnone -Werror -Wall -emit-llvm -o - %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s
// RUN: %clang_cc1 -fclang-abi-compat=latest -triple aarch64 -target-feature +sme -target-feature +sme2 -target-feature +sme-i16i64 -target-feature +sme-f64f64 -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 -fclang-abi-compat=latest -DSVE_OVERLOADED_FORMS -triple aarch64 -target-feature +sme -target-feature +sme2 -target-feature +sme-i16i64 -target-feature +sme-f64f64 -disable-O0-optnone -Werror -Wall -emit-llvm -o - %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s
// RUN: %clang_cc1 -fclang-abi-compat=latest -DSVE_OVERLOADED_FORMS -triple aarch64 -target-feature +sme -target-feature +sme2 -target-feature +sme-i16i64 -target-feature +sme-f64f64 -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 -fclang-abi-compat=latest -triple aarch64 -target-feature +sme -target-feature +sme2 -target-feature +sme-i16i64 -target-feature +sme-f64f64 -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,A5) A1##A3##A5
#else
#define SVE_ACLE_FUNC(A1,A2,A3,A4,A5) A1##A2##A3##A4##A5
#endif

//
// Single-Multi
//

// x2
// CHECK-LABEL: @test_svadd_write_single2_s32(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.single.za.vg1x2.nxv4i32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x i32> [[ZN_COERCE0:%.*]], <vscale x 4 x i32> [[ZN_COERCE1:%.*]], <vscale x 4 x i32> [[ZM:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z28test_svadd_write_single2_s32j11svint32x2_tu11__SVInt32_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.single.za.vg1x2.nxv4i32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x i32> [[ZN_COERCE0:%.*]], <vscale x 4 x i32> [[ZN_COERCE1:%.*]], <vscale x 4 x i32> [[ZM:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_write_single2_s32(uint32_t slice_base, svint32x2_t zn, svint32_t zm) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_write,_single,_za32,_s32,_vg1x2)(slice_base, zn, zm);
}

// CHECK-LABEL: @test_svadd_write_single2_u32(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.single.za.vg1x2.nxv4i32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x i32> [[ZN_COERCE0:%.*]], <vscale x 4 x i32> [[ZN_COERCE1:%.*]], <vscale x 4 x i32> [[ZM:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z28test_svadd_write_single2_u32j12svuint32x2_tu12__SVUint32_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.single.za.vg1x2.nxv4i32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x i32> [[ZN_COERCE0:%.*]], <vscale x 4 x i32> [[ZN_COERCE1:%.*]], <vscale x 4 x i32> [[ZM:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_write_single2_u32(uint32_t slice_base, svuint32x2_t zn, svuint32_t zm) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_write,_single,_za32,_u32,_vg1x2)(slice_base, zn, zm);
}

// CHECK-LABEL: @test_svadd_write_single2_s64(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.single.za.vg1x2.nxv2i64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x i64> [[ZN_COERCE0:%.*]], <vscale x 2 x i64> [[ZN_COERCE1:%.*]], <vscale x 2 x i64> [[ZM:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z28test_svadd_write_single2_s64j11svint64x2_tu11__SVInt64_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.single.za.vg1x2.nxv2i64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x i64> [[ZN_COERCE0:%.*]], <vscale x 2 x i64> [[ZN_COERCE1:%.*]], <vscale x 2 x i64> [[ZM:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_write_single2_s64(uint32_t slice_base, svint64x2_t zn, svint64_t zm) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_write,_single,_za64,_s64,_vg1x2)(slice_base, zn, zm);
}

// CHECK-LABEL: @test_svadd_write_single2_u64(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.single.za.vg1x2.nxv2i64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x i64> [[ZN_COERCE0:%.*]], <vscale x 2 x i64> [[ZN_COERCE1:%.*]], <vscale x 2 x i64> [[ZM:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z28test_svadd_write_single2_u64j12svuint64x2_tu12__SVUint64_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.single.za.vg1x2.nxv2i64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x i64> [[ZN_COERCE0:%.*]], <vscale x 2 x i64> [[ZN_COERCE1:%.*]], <vscale x 2 x i64> [[ZM:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_write_single2_u64(uint32_t slice_base, svuint64x2_t zn, svuint64_t zm) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_write,_single,_za64,_u64,_vg1x2)(slice_base, zn, zm);
}

// x4

// CHECK-LABEL: @test_svadd_write_single4_s32(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.single.za.vg1x4.nxv4i32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x i32> [[ZN_COERCE0:%.*]], <vscale x 4 x i32> [[ZN_COERCE1:%.*]], <vscale x 4 x i32> [[ZN_COERCE2:%.*]], <vscale x 4 x i32> [[ZN_COERCE3:%.*]], <vscale x 4 x i32> [[ZM:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z28test_svadd_write_single4_s32j11svint32x4_tu11__SVInt32_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.single.za.vg1x4.nxv4i32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x i32> [[ZN_COERCE0:%.*]], <vscale x 4 x i32> [[ZN_COERCE1:%.*]], <vscale x 4 x i32> [[ZN_COERCE2:%.*]], <vscale x 4 x i32> [[ZN_COERCE3:%.*]], <vscale x 4 x i32> [[ZM:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_write_single4_s32(uint32_t slice_base, svint32x4_t zn, svint32_t zm) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_write,_single,_za32,_s32,_vg1x4)(slice_base, zn, zm);
}

// CHECK-LABEL: @test_svadd_write_single4_u32(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.single.za.vg1x4.nxv4i32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x i32> [[ZN_COERCE0:%.*]], <vscale x 4 x i32> [[ZN_COERCE1:%.*]], <vscale x 4 x i32> [[ZN_COERCE2:%.*]], <vscale x 4 x i32> [[ZN_COERCE3:%.*]], <vscale x 4 x i32> [[ZM:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z28test_svadd_write_single4_u32j12svuint32x4_tu12__SVUint32_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.single.za.vg1x4.nxv4i32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x i32> [[ZN_COERCE0:%.*]], <vscale x 4 x i32> [[ZN_COERCE1:%.*]], <vscale x 4 x i32> [[ZN_COERCE2:%.*]], <vscale x 4 x i32> [[ZN_COERCE3:%.*]], <vscale x 4 x i32> [[ZM:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_write_single4_u32(uint32_t slice_base, svuint32x4_t zn, svuint32_t zm) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_write,_single,_za32,_u32,_vg1x4)(slice_base, zn, zm);
}

// CHECK-LABEL: @test_svadd_write_single4_s64(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.single.za.vg1x4.nxv2i64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x i64> [[ZN_COERCE0:%.*]], <vscale x 2 x i64> [[ZN_COERCE1:%.*]], <vscale x 2 x i64> [[ZN_COERCE2:%.*]], <vscale x 2 x i64> [[ZN_COERCE3:%.*]], <vscale x 2 x i64> [[ZM:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z28test_svadd_write_single4_s64j11svint64x4_tu11__SVInt64_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.single.za.vg1x4.nxv2i64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x i64> [[ZN_COERCE0:%.*]], <vscale x 2 x i64> [[ZN_COERCE1:%.*]], <vscale x 2 x i64> [[ZN_COERCE2:%.*]], <vscale x 2 x i64> [[ZN_COERCE3:%.*]], <vscale x 2 x i64> [[ZM:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_write_single4_s64(uint32_t slice_base, svint64x4_t zn, svint64_t zm) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_write,_single,_za64,_s64,_vg1x4)(slice_base, zn, zm);
}

// CHECK-LABEL: @test_svadd_write_single4_u64(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.single.za.vg1x4.nxv2i64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x i64> [[ZN_COERCE0:%.*]], <vscale x 2 x i64> [[ZN_COERCE1:%.*]], <vscale x 2 x i64> [[ZN_COERCE2:%.*]], <vscale x 2 x i64> [[ZN_COERCE3:%.*]], <vscale x 2 x i64> [[ZM:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z28test_svadd_write_single4_u64j12svuint64x4_tu12__SVUint64_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.single.za.vg1x4.nxv2i64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x i64> [[ZN_COERCE0:%.*]], <vscale x 2 x i64> [[ZN_COERCE1:%.*]], <vscale x 2 x i64> [[ZN_COERCE2:%.*]], <vscale x 2 x i64> [[ZN_COERCE3:%.*]], <vscale x 2 x i64> [[ZM:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_write_single4_u64(uint32_t slice_base, svuint64x4_t zn, svuint64_t zm) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_write,_single,_za64,_u64,_vg1x4)(slice_base, zn, zm);
}

//
// Multi-Multi
//

// x2

// CHECK-LABEL: @test_svadd_write_multi2_s32(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.za.vg1x2.nxv4i32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x i32> [[ZN_COERCE0:%.*]], <vscale x 4 x i32> [[ZN_COERCE1:%.*]], <vscale x 4 x i32> [[ZM_COERCE0:%.*]], <vscale x 4 x i32> [[ZM_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svadd_write_multi2_s32j11svint32x2_tS_(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.za.vg1x2.nxv4i32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x i32> [[ZN_COERCE0:%.*]], <vscale x 4 x i32> [[ZN_COERCE1:%.*]], <vscale x 4 x i32> [[ZM_COERCE0:%.*]], <vscale x 4 x i32> [[ZM_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_write_multi2_s32(uint32_t slice_base, svint32x2_t zn, svint32x2_t zm) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_write,,_za32,_s32,_vg1x2)(slice_base, zn, zm);
}

// CHECK-LABEL: @test_svadd_write_multi2_u32(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.za.vg1x2.nxv4i32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x i32> [[ZN_COERCE0:%.*]], <vscale x 4 x i32> [[ZN_COERCE1:%.*]], <vscale x 4 x i32> [[ZM_COERCE0:%.*]], <vscale x 4 x i32> [[ZM_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svadd_write_multi2_u32j12svuint32x2_tS_(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.za.vg1x2.nxv4i32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x i32> [[ZN_COERCE0:%.*]], <vscale x 4 x i32> [[ZN_COERCE1:%.*]], <vscale x 4 x i32> [[ZM_COERCE0:%.*]], <vscale x 4 x i32> [[ZM_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_write_multi2_u32(uint32_t slice_base, svuint32x2_t zn, svuint32x2_t zm) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_write,,_za32,_u32,_vg1x2)(slice_base, zn, zm);
}

// CHECK-LABEL: @test_svadd_write_multi2_s64(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.za.vg1x2.nxv2i64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x i64> [[ZN_COERCE0:%.*]], <vscale x 2 x i64> [[ZN_COERCE1:%.*]], <vscale x 2 x i64> [[ZM_COERCE0:%.*]], <vscale x 2 x i64> [[ZM_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svadd_write_multi2_s64j11svint64x2_tS_(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.za.vg1x2.nxv2i64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x i64> [[ZN_COERCE0:%.*]], <vscale x 2 x i64> [[ZN_COERCE1:%.*]], <vscale x 2 x i64> [[ZM_COERCE0:%.*]], <vscale x 2 x i64> [[ZM_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_write_multi2_s64(uint32_t slice_base, svint64x2_t zn, svint64x2_t zm) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_write,,_za64,_s64,_vg1x2)(slice_base, zn, zm);
}

// CHECK-LABEL: @test_svadd_write_multi2_u64(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.za.vg1x2.nxv2i64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x i64> [[ZN_COERCE0:%.*]], <vscale x 2 x i64> [[ZN_COERCE1:%.*]], <vscale x 2 x i64> [[ZM_COERCE0:%.*]], <vscale x 2 x i64> [[ZM_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svadd_write_multi2_u64j12svuint64x2_tS_(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.za.vg1x2.nxv2i64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x i64> [[ZN_COERCE0:%.*]], <vscale x 2 x i64> [[ZN_COERCE1:%.*]], <vscale x 2 x i64> [[ZM_COERCE0:%.*]], <vscale x 2 x i64> [[ZM_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_write_multi2_u64(uint32_t slice_base, svuint64x2_t zn, svuint64x2_t zm) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_write,,_za64,_u64,_vg1x2)(slice_base, zn, zm);
}

// x4

// CHECK-LABEL: @test_svadd_write_multi4_s32(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.za.vg1x4.nxv4i32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x i32> [[ZN_COERCE0:%.*]], <vscale x 4 x i32> [[ZN_COERCE1:%.*]], <vscale x 4 x i32> [[ZN_COERCE2:%.*]], <vscale x 4 x i32> [[ZN_COERCE3:%.*]], <vscale x 4 x i32> [[ZM_COERCE0:%.*]], <vscale x 4 x i32> [[ZM_COERCE1:%.*]], <vscale x 4 x i32> [[ZM_COERCE2:%.*]], <vscale x 4 x i32> [[ZM_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svadd_write_multi4_s32j11svint32x4_tS_(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.za.vg1x4.nxv4i32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x i32> [[ZN_COERCE0:%.*]], <vscale x 4 x i32> [[ZN_COERCE1:%.*]], <vscale x 4 x i32> [[ZN_COERCE2:%.*]], <vscale x 4 x i32> [[ZN_COERCE3:%.*]], <vscale x 4 x i32> [[ZM_COERCE0:%.*]], <vscale x 4 x i32> [[ZM_COERCE1:%.*]], <vscale x 4 x i32> [[ZM_COERCE2:%.*]], <vscale x 4 x i32> [[ZM_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_write_multi4_s32(uint32_t slice_base, svint32x4_t zn, svint32x4_t zm) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_write,,_za32,_s32,_vg1x4)(slice_base, zn, zm);
}

// CHECK-LABEL: @test_svadd_write_multi4_u32(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.za.vg1x4.nxv4i32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x i32> [[ZN_COERCE0:%.*]], <vscale x 4 x i32> [[ZN_COERCE1:%.*]], <vscale x 4 x i32> [[ZN_COERCE2:%.*]], <vscale x 4 x i32> [[ZN_COERCE3:%.*]], <vscale x 4 x i32> [[ZM_COERCE0:%.*]], <vscale x 4 x i32> [[ZM_COERCE1:%.*]], <vscale x 4 x i32> [[ZM_COERCE2:%.*]], <vscale x 4 x i32> [[ZM_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svadd_write_multi4_u32j12svuint32x4_tS_(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.za.vg1x4.nxv4i32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x i32> [[ZN_COERCE0:%.*]], <vscale x 4 x i32> [[ZN_COERCE1:%.*]], <vscale x 4 x i32> [[ZN_COERCE2:%.*]], <vscale x 4 x i32> [[ZN_COERCE3:%.*]], <vscale x 4 x i32> [[ZM_COERCE0:%.*]], <vscale x 4 x i32> [[ZM_COERCE1:%.*]], <vscale x 4 x i32> [[ZM_COERCE2:%.*]], <vscale x 4 x i32> [[ZM_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_write_multi4_u32(uint32_t slice_base, svuint32x4_t zn, svuint32x4_t zm) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_write,,_za32,_u32,_vg1x4)(slice_base, zn, zm);
}

// CHECK-LABEL: @test_svadd_write_multi4_s64(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.za.vg1x4.nxv2i64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x i64> [[ZN_COERCE0:%.*]], <vscale x 2 x i64> [[ZN_COERCE1:%.*]], <vscale x 2 x i64> [[ZN_COERCE2:%.*]], <vscale x 2 x i64> [[ZN_COERCE3:%.*]], <vscale x 2 x i64> [[ZM_COERCE0:%.*]], <vscale x 2 x i64> [[ZM_COERCE1:%.*]], <vscale x 2 x i64> [[ZM_COERCE2:%.*]], <vscale x 2 x i64> [[ZM_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svadd_write_multi4_s64j11svint64x4_tS_(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.za.vg1x4.nxv2i64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x i64> [[ZN_COERCE0:%.*]], <vscale x 2 x i64> [[ZN_COERCE1:%.*]], <vscale x 2 x i64> [[ZN_COERCE2:%.*]], <vscale x 2 x i64> [[ZN_COERCE3:%.*]], <vscale x 2 x i64> [[ZM_COERCE0:%.*]], <vscale x 2 x i64> [[ZM_COERCE1:%.*]], <vscale x 2 x i64> [[ZM_COERCE2:%.*]], <vscale x 2 x i64> [[ZM_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_write_multi4_s64(uint32_t slice_base, svint64x4_t zn, svint64x4_t zm) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_write,,_za64,_s64,_vg1x4)(slice_base, zn, zm);
}

// CHECK-LABEL: @test_svadd_write_multi4_u64(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.za.vg1x4.nxv2i64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x i64> [[ZN_COERCE0:%.*]], <vscale x 2 x i64> [[ZN_COERCE1:%.*]], <vscale x 2 x i64> [[ZN_COERCE2:%.*]], <vscale x 2 x i64> [[ZN_COERCE3:%.*]], <vscale x 2 x i64> [[ZM_COERCE0:%.*]], <vscale x 2 x i64> [[ZM_COERCE1:%.*]], <vscale x 2 x i64> [[ZM_COERCE2:%.*]], <vscale x 2 x i64> [[ZM_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z27test_svadd_write_multi4_u64j12svuint64x4_tS_(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.write.za.vg1x4.nxv2i64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x i64> [[ZN_COERCE0:%.*]], <vscale x 2 x i64> [[ZN_COERCE1:%.*]], <vscale x 2 x i64> [[ZN_COERCE2:%.*]], <vscale x 2 x i64> [[ZN_COERCE3:%.*]], <vscale x 2 x i64> [[ZM_COERCE0:%.*]], <vscale x 2 x i64> [[ZM_COERCE1:%.*]], <vscale x 2 x i64> [[ZM_COERCE2:%.*]], <vscale x 2 x i64> [[ZM_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_write_multi4_u64(uint32_t slice_base, svuint64x4_t zn, svuint64x4_t zm) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_write,,_za64,_u64,_vg1x4)(slice_base, zn, zm);
}

//
// Accumulate to ZA
//

// x2

// CHECK-LABEL: @test_svadd_za32_vg1x2_f32(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.za32.vg1x2.nxv4f32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x float> [[ZN_COERCE0:%.*]], <vscale x 4 x float> [[ZN_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z25test_svadd_za32_vg1x2_f32j13svfloat32x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.za32.vg1x2.nxv4f32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x float> [[ZN_COERCE0:%.*]], <vscale x 4 x float> [[ZN_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_za32_vg1x2_f32(uint32_t slice_base, svfloat32x2_t zn) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_za32,,_f32,,_vg1x2)(slice_base, zn);
}

// CHECK-LABEL: @test_svadd_za32_vg1x2_s32(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.za32.vg1x2.nxv4i32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x i32> [[ZN_COERCE0:%.*]], <vscale x 4 x i32> [[ZN_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z25test_svadd_za32_vg1x2_s32j11svint32x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.za32.vg1x2.nxv4i32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x i32> [[ZN_COERCE0:%.*]], <vscale x 4 x i32> [[ZN_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_za32_vg1x2_s32(uint32_t slice_base, svint32x2_t zn) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_za32,,_s32,,_vg1x2)(slice_base , zn);
}

// CHECK-LABEL: @test_svadd_za32_vg1x2_u32(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.za32.vg1x2.nxv4i32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x i32> [[ZN_COERCE0:%.*]], <vscale x 4 x i32> [[ZN_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z25test_svadd_za32_vg1x2_u32j12svuint32x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.za32.vg1x2.nxv4i32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x i32> [[ZN_COERCE0:%.*]], <vscale x 4 x i32> [[ZN_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_za32_vg1x2_u32(uint32_t slice_base, svuint32x2_t zn) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_za32,,_u32,,_vg1x2)(slice_base, zn);
}

// CHECK-LABEL: @test_svadd_za64_vg1x2_f64(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.za64.vg1x2.nxv2f64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x double> [[ZN_COERCE0:%.*]], <vscale x 2 x double> [[ZN_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z25test_svadd_za64_vg1x2_f64j13svfloat64x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.za64.vg1x2.nxv2f64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x double> [[ZN_COERCE0:%.*]], <vscale x 2 x double> [[ZN_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_za64_vg1x2_f64(uint32_t slice_base, svfloat64x2_t zn) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_za64,,_f64,,_vg1x2)(slice_base, zn);
}

// CHECK-LABEL: @test_svadd_za64_vg1x2_s64(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.za64.vg1x2.nxv2i64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x i64> [[ZN_COERCE0:%.*]], <vscale x 2 x i64> [[ZN_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z25test_svadd_za64_vg1x2_s64j11svint64x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.za64.vg1x2.nxv2i64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x i64> [[ZN_COERCE0:%.*]], <vscale x 2 x i64> [[ZN_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_za64_vg1x2_s64(uint32_t slice_base, svint64x2_t zn) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_za64,,_s64,,_vg1x2)(slice_base, zn);
}

// CHECK-LABEL: @test_svadd_za64_vg1x2_u64(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.za64.vg1x2.nxv2i64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x i64> [[ZN_COERCE0:%.*]], <vscale x 2 x i64> [[ZN_COERCE1:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z25test_svadd_za64_vg1x2_u64j12svuint64x2_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.za64.vg1x2.nxv2i64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x i64> [[ZN_COERCE0:%.*]], <vscale x 2 x i64> [[ZN_COERCE1:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_za64_vg1x2_u64(uint32_t slice_base, svuint64x2_t zn) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_za64,,_u64,,_vg1x2)(slice_base, zn);
}

// x4

// CHECK-LABEL: @test_svadd_za32_vg1x4_f32(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.za32.vg1x4.nxv4f32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x float> [[ZN_COERCE0:%.*]], <vscale x 4 x float> [[ZN_COERCE1:%.*]], <vscale x 4 x float> [[ZN_COERCE2:%.*]], <vscale x 4 x float> [[ZN_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z25test_svadd_za32_vg1x4_f32j13svfloat32x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.za32.vg1x4.nxv4f32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x float> [[ZN_COERCE0:%.*]], <vscale x 4 x float> [[ZN_COERCE1:%.*]], <vscale x 4 x float> [[ZN_COERCE2:%.*]], <vscale x 4 x float> [[ZN_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_za32_vg1x4_f32(uint32_t slice_base, svfloat32x4_t zn) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_za32,,_f32,,_vg1x4)(slice_base, zn);
}

// CHECK-LABEL: @test_svadd_za32_vg1x4_s32(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.za32.vg1x4.nxv4i32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x i32> [[ZN_COERCE0:%.*]], <vscale x 4 x i32> [[ZN_COERCE1:%.*]], <vscale x 4 x i32> [[ZN_COERCE2:%.*]], <vscale x 4 x i32> [[ZN_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z25test_svadd_za32_vg1x4_s32j11svint32x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.za32.vg1x4.nxv4i32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x i32> [[ZN_COERCE0:%.*]], <vscale x 4 x i32> [[ZN_COERCE1:%.*]], <vscale x 4 x i32> [[ZN_COERCE2:%.*]], <vscale x 4 x i32> [[ZN_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_za32_vg1x4_s32(uint32_t slice_base, svint32x4_t zn) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_za32,,_s32,,_vg1x4)(slice_base, zn);
}

// CHECK-LABEL: @test_svadd_za32_vg1x4_u32(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.za32.vg1x4.nxv4i32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x i32> [[ZN_COERCE0:%.*]], <vscale x 4 x i32> [[ZN_COERCE1:%.*]], <vscale x 4 x i32> [[ZN_COERCE2:%.*]], <vscale x 4 x i32> [[ZN_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z25test_svadd_za32_vg1x4_u32j12svuint32x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.za32.vg1x4.nxv4i32(i32 [[SLICE_BASE:%.*]], <vscale x 4 x i32> [[ZN_COERCE0:%.*]], <vscale x 4 x i32> [[ZN_COERCE1:%.*]], <vscale x 4 x i32> [[ZN_COERCE2:%.*]], <vscale x 4 x i32> [[ZN_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_za32_vg1x4_u32(uint32_t slice_base, svuint32x4_t zn) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_za32,,_u32,,_vg1x4)(slice_base, zn);
}

// CHECK-LABEL: @test_svadd_za64_vg1x4_f64(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.za64.vg1x4.nxv2f64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x double> [[ZN_COERCE0:%.*]], <vscale x 2 x double> [[ZN_COERCE1:%.*]], <vscale x 2 x double> [[ZN_COERCE2:%.*]], <vscale x 2 x double> [[ZN_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z25test_svadd_za64_vg1x4_f64j13svfloat64x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.za64.vg1x4.nxv2f64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x double> [[ZN_COERCE0:%.*]], <vscale x 2 x double> [[ZN_COERCE1:%.*]], <vscale x 2 x double> [[ZN_COERCE2:%.*]], <vscale x 2 x double> [[ZN_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_za64_vg1x4_f64(uint32_t slice_base, svfloat64x4_t zn) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_za64,,_f64,,_vg1x4)(slice_base, zn);
}

// CHECK-LABEL: @test_svadd_za64_vg1x4_s64(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.za64.vg1x4.nxv2i64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x i64> [[ZN_COERCE0:%.*]], <vscale x 2 x i64> [[ZN_COERCE1:%.*]], <vscale x 2 x i64> [[ZN_COERCE2:%.*]], <vscale x 2 x i64> [[ZN_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z25test_svadd_za64_vg1x4_s64j11svint64x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.za64.vg1x4.nxv2i64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x i64> [[ZN_COERCE0:%.*]], <vscale x 2 x i64> [[ZN_COERCE1:%.*]], <vscale x 2 x i64> [[ZN_COERCE2:%.*]], <vscale x 2 x i64> [[ZN_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_za64_vg1x4_s64(uint32_t slice_base, svint64x4_t zn) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_za64,,_s64,,_vg1x4)(slice_base, zn);
}

// CHECK-LABEL: @test_svadd_za64_vg1x4_u64(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.za64.vg1x4.nxv2i64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x i64> [[ZN_COERCE0:%.*]], <vscale x 2 x i64> [[ZN_COERCE1:%.*]], <vscale x 2 x i64> [[ZN_COERCE2:%.*]], <vscale x 2 x i64> [[ZN_COERCE3:%.*]])
// CHECK-NEXT:    ret void
//
// CPP-CHECK-LABEL: @_Z25test_svadd_za64_vg1x4_u64j12svuint64x4_t(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    tail call void @llvm.aarch64.sme.add.za64.vg1x4.nxv2i64(i32 [[SLICE_BASE:%.*]], <vscale x 2 x i64> [[ZN_COERCE0:%.*]], <vscale x 2 x i64> [[ZN_COERCE1:%.*]], <vscale x 2 x i64> [[ZN_COERCE2:%.*]], <vscale x 2 x i64> [[ZN_COERCE3:%.*]])
// CPP-CHECK-NEXT:    ret void
//
void test_svadd_za64_vg1x4_u64(uint32_t slice_base, svuint64x4_t zn) __arm_streaming __arm_inout("za") {
  SVE_ACLE_FUNC(svadd_za64,,_u64,,_vg1x4)(slice_base, zn);
}