Codebase Browser
llvm
Go to App

llvm/clang/test/CodeGen/aarch64-sme2-intrinsics/acle_sme2_read.c 

// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
// REQUIRES: aarch64-registered-target

// RUN: %clang_cc1 -triple aarch64 -target-feature +bf16 -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 +bf16 -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 +bf16 -target-feature +sme -target-feature +sme2 -S -disable-O0-optnone -Werror -Wall -o /dev/null %s

#include <arm_sme.h>

// CHECK-LABEL: @test_svread_ver_za8_u8_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.read.ver.vg2.nxv16i8(i32 0, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_ver_za8_u8_vg2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.read.ver.vg2.nxv16i8(i32 0, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]]
//
svuint8x2_t test_svread_ver_za8_u8_vg2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_ver_za8_u8_vg2(0, base);
}

// CHECK-LABEL: @test_svread_ver_za8_s8_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.read.ver.vg2.nxv16i8(i32 0, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_ver_za8_s8_vg2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.read.ver.vg2.nxv16i8(i32 0, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]]
//
svint8x2_t test_svread_ver_za8_s8_vg2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_ver_za8_s8_vg2(0, base);
}

// CHECK-LABEL: @test_svread_hor_za8_u8_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.read.hor.vg2.nxv16i8(i32 0, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_hor_za8_u8_vg2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.read.hor.vg2.nxv16i8(i32 0, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]]
//
svuint8x2_t test_svread_hor_za8_u8_vg2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_hor_za8_u8_vg2(0, base);
}

// CHECK-LABEL: @test_svread_hor_za8_s8_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.read.hor.vg2.nxv16i8(i32 0, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_hor_za8_s8_vg2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.read.hor.vg2.nxv16i8(i32 0, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]]
//
svint8x2_t test_svread_hor_za8_s8_vg2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_hor_za8_s8_vg2(0, base);
}

// CHECK-LABEL: @test_svread_hor_za8_u8_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.read.hor.vg4.nxv16i8(i32 0, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_hor_za8_u8_vg4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.read.hor.vg4.nxv16i8(i32 0, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]]
//
svuint8x4_t test_svread_hor_za8_u8_vg4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_hor_za8_u8_vg4(0, base);
}

// CHECK-LABEL: @test_svread_hor_za8_s8_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.read.hor.vg4.nxv16i8(i32 0, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_hor_za8_s8_vg4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.read.hor.vg4.nxv16i8(i32 0, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]]
//
svint8x4_t test_svread_hor_za8_s8_vg4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_hor_za8_s8_vg4(0, base);
}

// CHECK-LABEL: @test_svread_ver_za8_u8_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.read.ver.vg4.nxv16i8(i32 0, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_ver_za8_u8_vg4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.read.ver.vg4.nxv16i8(i32 0, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]]
//
svuint8x4_t test_svread_ver_za8_u8_vg4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_ver_za8_u8_vg4(0, base);
}

// CHECK-LABEL: @test_svread_ver_za8_s8_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.read.ver.vg4.nxv16i8(i32 0, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_ver_za8_s8_vg4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.read.ver.vg4.nxv16i8(i32 0, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]]
//
svint8x4_t test_svread_ver_za8_s8_vg4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_ver_za8_s8_vg4(0, base);
}

// CHECK-LABEL: @test_svread_hor_za16_u16_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.read.hor.vg2.nxv8i16(i32 1, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_hor_za16_u16_vg2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.read.hor.vg2.nxv8i16(i32 1, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]]
//
svuint16x2_t test_svread_hor_za16_u16_vg2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_hor_za16_u16_vg2(1, base);
}

// CHECK-LABEL: @test_svread_hor_za16_bf16_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.aarch64.sme.read.hor.vg2.nxv8bf16(i32 1, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z29test_svread_hor_za16_bf16_vg2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.aarch64.sme.read.hor.vg2.nxv8bf16(i32 1, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
//
svbfloat16x2_t test_svread_hor_za16_bf16_vg2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_hor_za16_bf16_vg2(1, base);
}

// CHECK-LABEL: @test_svread_hor_za16_f16_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sme.read.hor.vg2.nxv8f16(i32 1, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 8 x half>, <vscale x 8 x half> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_hor_za16_f16_vg2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sme.read.hor.vg2.nxv8f16(i32 1, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 8 x half>, <vscale x 8 x half> } [[TMP0]]
//
svfloat16x2_t test_svread_hor_za16_f16_vg2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_hor_za16_f16_vg2(1, base);
}

// CHECK-LABEL: @test_svread_hor_za16_s16_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.read.hor.vg2.nxv8i16(i32 1, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_hor_za16_s16_vg2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.read.hor.vg2.nxv8i16(i32 1, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]]
//
svint16x2_t test_svread_hor_za16_s16_vg2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_hor_za16_s16_vg2(1, base);
}

// CHECK-LABEL: @test_svread_ver_za16_u16_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.read.ver.vg2.nxv8i16(i32 1, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_ver_za16_u16_vg2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.read.ver.vg2.nxv8i16(i32 1, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]]
//
svuint16x2_t test_svread_ver_za16_u16_vg2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_ver_za16_u16_vg2(1, base);
}

// CHECK-LABEL: @test_svread_ver_za16_bf16_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.aarch64.sme.read.ver.vg2.nxv8bf16(i32 1, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z29test_svread_ver_za16_bf16_vg2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.aarch64.sme.read.ver.vg2.nxv8bf16(i32 1, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
//
svbfloat16x2_t test_svread_ver_za16_bf16_vg2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_ver_za16_bf16_vg2(1, base);
}

// CHECK-LABEL: @test_svread_ver_za16_f16_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sme.read.ver.vg2.nxv8f16(i32 1, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 8 x half>, <vscale x 8 x half> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_ver_za16_f16_vg2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sme.read.ver.vg2.nxv8f16(i32 1, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 8 x half>, <vscale x 8 x half> } [[TMP0]]
//
svfloat16x2_t test_svread_ver_za16_f16_vg2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_ver_za16_f16_vg2(1, base);
}

// CHECK-LABEL: @test_svread_ver_za16_s16_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.read.ver.vg2.nxv8i16(i32 1, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_ver_za16_s16_vg2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.read.ver.vg2.nxv8i16(i32 1, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]]
//
svint16x2_t test_svread_ver_za16_s16_vg2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_ver_za16_s16_vg2(1, base);
}

// CHECK-LABEL: @test_svread_hor_za16_u16_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.read.hor.vg4.nxv8i16(i32 1, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_hor_za16_u16_vg4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.read.hor.vg4.nxv8i16(i32 1, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]]
//
svuint16x4_t test_svread_hor_za16_u16_vg4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_hor_za16_u16_vg4(1, base);
}

// CHECK-LABEL: @test_svread_hor_za16_bf16_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.aarch64.sme.read.hor.vg4.nxv8bf16(i32 1, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z29test_svread_hor_za16_bf16_vg4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.aarch64.sme.read.hor.vg4.nxv8bf16(i32 1, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
//
svbfloat16x4_t test_svread_hor_za16_bf16_vg4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_hor_za16_bf16_vg4(1, base);
}

// CHECK-LABEL: @test_svread_hor_za16_f16_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sme.read.hor.vg4.nxv8f16(i32 1, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_hor_za16_f16_vg4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sme.read.hor.vg4.nxv8f16(i32 1, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half> } [[TMP0]]
//
svfloat16x4_t test_svread_hor_za16_f16_vg4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_hor_za16_f16_vg4(1, base);
}

// CHECK-LABEL: @test_svread_hor_za16_s16_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.read.hor.vg4.nxv8i16(i32 1, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_hor_za16_s16_vg4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.read.hor.vg4.nxv8i16(i32 1, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]]
//
svint16x4_t test_svread_hor_za16_s16_vg4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_hor_za16_s16_vg4(1, base);
}

// CHECK-LABEL: @test_svread_ver_za16_u16_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.read.ver.vg4.nxv8i16(i32 1, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_ver_za16_u16_vg4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.read.ver.vg4.nxv8i16(i32 1, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]]
//
svuint16x4_t test_svread_ver_za16_u16_vg4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_ver_za16_u16_vg4(1, base);
}

// CHECK-LABEL: @test_svread_ver_za16_bf16_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.aarch64.sme.read.ver.vg4.nxv8bf16(i32 1, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z29test_svread_ver_za16_bf16_vg4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.aarch64.sme.read.ver.vg4.nxv8bf16(i32 1, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
//
svbfloat16x4_t test_svread_ver_za16_bf16_vg4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_ver_za16_bf16_vg4(1, base);
}

// CHECK-LABEL: @test_svread_ver_za16_f16_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sme.read.ver.vg4.nxv8f16(i32 1, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_ver_za16_f16_vg4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sme.read.ver.vg4.nxv8f16(i32 1, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half> } [[TMP0]]
//
svfloat16x4_t test_svread_ver_za16_f16_vg4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_ver_za16_f16_vg4(1, base);
}

// CHECK-LABEL: @test_svread_ver_za16_s16_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.read.ver.vg4.nxv8i16(i32 1, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_ver_za16_s16_vg4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.read.ver.vg4.nxv8i16(i32 1, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]]
//
svint16x4_t test_svread_ver_za16_s16_vg4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_ver_za16_s16_vg4(1, base);
}

// CHECK-LABEL: @test_svread_hor_za32_u32_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.read.hor.vg2.nxv4i32(i32 3, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_hor_za32_u32_vg2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.read.hor.vg2.nxv4i32(i32 3, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]]
//
svuint32x2_t test_svread_hor_za32_u32_vg2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_hor_za32_u32_vg2(3, base);
}

// CHECK-LABEL: @test_svread_hor_za32_f32_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sme.read.hor.vg2.nxv4f32(i32 3, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 4 x float>, <vscale x 4 x float> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_hor_za32_f32_vg2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sme.read.hor.vg2.nxv4f32(i32 3, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 4 x float>, <vscale x 4 x float> } [[TMP0]]
//
svfloat32x2_t test_svread_hor_za32_f32_vg2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_hor_za32_f32_vg2(3, base);
}

// CHECK-LABEL: @test_svread_hor_za32_s32_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.read.hor.vg2.nxv4i32(i32 3, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_hor_za32_s32_vg2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.read.hor.vg2.nxv4i32(i32 3, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]]
//
svint32x2_t test_svread_hor_za32_s32_vg2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_hor_za32_s32_vg2(3, base);
}

// CHECK-LABEL: @test_svread_ver_za32_u32_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.read.ver.vg2.nxv4i32(i32 3, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_ver_za32_u32_vg2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.read.ver.vg2.nxv4i32(i32 3, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]]
//
svuint32x2_t test_svread_ver_za32_u32_vg2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_ver_za32_u32_vg2(3, base);
}

// CHECK-LABEL: @test_svread_ver_za32_f32_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sme.read.ver.vg2.nxv4f32(i32 3, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 4 x float>, <vscale x 4 x float> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_ver_za32_f32_vg2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sme.read.ver.vg2.nxv4f32(i32 3, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 4 x float>, <vscale x 4 x float> } [[TMP0]]
//
svfloat32x2_t test_svread_ver_za32_f32_vg2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_ver_za32_f32_vg2(3, base);
}

// CHECK-LABEL: @test_svread_ver_za32_s32_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.read.ver.vg2.nxv4i32(i32 3, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_ver_za32_s32_vg2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.read.ver.vg2.nxv4i32(i32 3, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]]
//
svint32x2_t test_svread_ver_za32_s32_vg2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_ver_za32_s32_vg2(3, base);
}

// CHECK-LABEL: @test_svread_hor_za32_u32_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.read.hor.vg4.nxv4i32(i32 3, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_hor_za32_u32_vg4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.read.hor.vg4.nxv4i32(i32 3, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]]
//
svuint32x4_t test_svread_hor_za32_u32_vg4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_hor_za32_u32_vg4(3, base);
}

// CHECK-LABEL: @test_svread_hor_za32_f32_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sme.read.hor.vg4.nxv4f32(i32 3, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_hor_za32_f32_vg4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sme.read.hor.vg4.nxv4f32(i32 3, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float> } [[TMP0]]
//
svfloat32x4_t test_svread_hor_za32_f32_vg4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_hor_za32_f32_vg4(3, base);
}

// CHECK-LABEL: @test_svread_hor_za32_s32_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.read.hor.vg4.nxv4i32(i32 3, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_hor_za32_s32_vg4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.read.hor.vg4.nxv4i32(i32 3, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]]
//
svint32x4_t test_svread_hor_za32_s32_vg4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_hor_za32_s32_vg4(3, base);
}

// CHECK-LABEL: @test_svread_ver_za32_u32_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.read.ver.vg4.nxv4i32(i32 3, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_ver_za32_u32_vg4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.read.ver.vg4.nxv4i32(i32 3, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]]
//
svuint32x4_t test_svread_ver_za32_u32_vg4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_ver_za32_u32_vg4(3, base);
}

// CHECK-LABEL: @test_svread_ver_za32_f32_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sme.read.ver.vg4.nxv4f32(i32 3, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_ver_za32_f32_vg4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sme.read.ver.vg4.nxv4f32(i32 3, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float> } [[TMP0]]
//
svfloat32x4_t test_svread_ver_za32_f32_vg4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_ver_za32_f32_vg4(3, base);
}

// CHECK-LABEL: @test_svread_ver_za32_s32_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.read.ver.vg4.nxv4i32(i32 3, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_ver_za32_s32_vg4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.read.ver.vg4.nxv4i32(i32 3, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]]
//
svint32x4_t test_svread_ver_za32_s32_vg4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_ver_za32_s32_vg4(3, base);
}

// CHECK-LABEL: @test_svread_hor_za64_u64_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sme.read.hor.vg2.nxv2i64(i32 7, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_hor_za64_u64_vg2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sme.read.hor.vg2.nxv2i64(i32 7, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64> } [[TMP0]]
//
svuint64x2_t test_svread_hor_za64_u64_vg2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_hor_za64_u64_vg2(7, base);
}

// CHECK-LABEL: @test_svread_hor_za64_f64_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.aarch64.sme.read.hor.vg2.nxv2f64(i32 7, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 2 x double>, <vscale x 2 x double> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_hor_za64_f64_vg2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.aarch64.sme.read.hor.vg2.nxv2f64(i32 7, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 2 x double>, <vscale x 2 x double> } [[TMP0]]
//
svfloat64x2_t test_svread_hor_za64_f64_vg2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_hor_za64_f64_vg2(7, base);
}

// CHECK-LABEL: @test_svread_hor_za64_s64_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sme.read.hor.vg2.nxv2i64(i32 7, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_hor_za64_s64_vg2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sme.read.hor.vg2.nxv2i64(i32 7, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64> } [[TMP0]]
//
svint64x2_t test_svread_hor_za64_s64_vg2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_hor_za64_s64_vg2(7, base);
}

// CHECK-LABEL: @test_svread_ver_za64_u64_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sme.read.ver.vg2.nxv2i64(i32 7, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_ver_za64_u64_vg2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sme.read.ver.vg2.nxv2i64(i32 7, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64> } [[TMP0]]
//
svuint64x2_t test_svread_ver_za64_u64_vg2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_ver_za64_u64_vg2(7, base);
}

// CHECK-LABEL: @test_svread_ver_za64_f64_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.aarch64.sme.read.ver.vg2.nxv2f64(i32 7, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 2 x double>, <vscale x 2 x double> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_ver_za64_f64_vg2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.aarch64.sme.read.ver.vg2.nxv2f64(i32 7, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 2 x double>, <vscale x 2 x double> } [[TMP0]]
//
svfloat64x2_t test_svread_ver_za64_f64_vg2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_ver_za64_f64_vg2(7, base);
}

//
// CHECK-LABEL: @test_svread_ver_za64_s64_vg2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sme.read.ver.vg2.nxv2i64(i32 7, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_ver_za64_s64_vg2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sme.read.ver.vg2.nxv2i64(i32 7, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64> } [[TMP0]]
//
svint64x2_t test_svread_ver_za64_s64_vg2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_ver_za64_s64_vg2(7, base);
}

// CHECK-LABEL: @test_svread_hor_za64_u64_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sme.read.hor.vg4.nxv2i64(i32 7, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_hor_za64_u64_vg4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sme.read.hor.vg4.nxv2i64(i32 7, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } [[TMP0]]
//
svuint64x4_t test_svread_hor_za64_u64_vg4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_hor_za64_u64_vg4(7, base);
}

// CHECK-LABEL: @test_svread_hor_za64_f64_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double> } @llvm.aarch64.sme.read.hor.vg4.nxv2f64(i32 7, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_hor_za64_f64_vg4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double> } @llvm.aarch64.sme.read.hor.vg4.nxv2f64(i32 7, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double> } [[TMP0]]
//
svfloat64x4_t test_svread_hor_za64_f64_vg4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_hor_za64_f64_vg4(7, base);
}

// CHECK-LABEL: @test_svread_hor_za64_s64_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sme.read.hor.vg4.nxv2i64(i32 7, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_hor_za64_s64_vg4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sme.read.hor.vg4.nxv2i64(i32 7, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } [[TMP0]]
//
svint64x4_t test_svread_hor_za64_s64_vg4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_hor_za64_s64_vg4(7, base);
}

// CHECK-LABEL: @test_svread_ver_za64_u64_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sme.read.ver.vg4.nxv2i64(i32 7, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_ver_za64_u64_vg4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sme.read.ver.vg4.nxv2i64(i32 7, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } [[TMP0]]
//
svuint64x4_t test_svread_ver_za64_u64_vg4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_ver_za64_u64_vg4(7, base);
}

// CHECK-LABEL: @test_svread_ver_za64_f64_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double> } @llvm.aarch64.sme.read.ver.vg4.nxv2f64(i32 7, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_ver_za64_f64_vg4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double> } @llvm.aarch64.sme.read.ver.vg4.nxv2f64(i32 7, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double> } [[TMP0]]
//
svfloat64x4_t test_svread_ver_za64_f64_vg4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_ver_za64_f64_vg4(7, base);
}

// CHECK-LABEL: @test_svread_ver_za64_s64_vg4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sme.read.ver.vg4.nxv2i64(i32 7, i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z28test_svread_ver_za64_s64_vg4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sme.read.ver.vg4.nxv2i64(i32 7, i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } [[TMP0]]
//
svint64x4_t test_svread_ver_za64_s64_vg4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_ver_za64_s64_vg4(7, base);
}

// CHECK-LABEL: @test_svread_za8_s8_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.read.vg1x2.nxv16i8(i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z24test_svread_za8_s8_vg1x2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.read.vg1x2.nxv16i8(i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]]
//
svint8x2_t test_svread_za8_s8_vg1x2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_za8_s8_vg1x2(base);
}

// CHECK-LABEL: @test_svread_za8_u8_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.read.vg1x2.nxv16i8(i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z24test_svread_za8_u8_vg1x2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.read.vg1x2.nxv16i8(i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]]
//
svuint8x2_t test_svread_za8_u8_vg1x2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_za8_u8_vg1x2(base);
}

// CHECK-LABEL: @test_svread_za16_s16_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.read.vg1x2.nxv8i16(i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_za16_s16_vg1x2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.read.vg1x2.nxv8i16(i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]]
//
svint16x2_t test_svread_za16_s16_vg1x2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_za16_s16_vg1x2(base);
}

// CHECK-LABEL: @test_svread_za16_u16_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.read.vg1x2.nxv8i16(i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_za16_u16_vg1x2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.read.vg1x2.nxv8i16(i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]]
//
svuint16x2_t test_svread_za16_u16_vg1x2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_za16_u16_vg1x2(base);
}

// CHECK-LABEL: @test_svread_za16_bf16_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.aarch64.sme.read.vg1x2.nxv8bf16(i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z27test_svread_za16_bf16_vg1x2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.aarch64.sme.read.vg1x2.nxv8bf16(i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
//
svbfloat16x2_t test_svread_za16_bf16_vg1x2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_za16_bf16_vg1x2(base);
}

// CHECK-LABEL: @test_svread_za16_f16_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sme.read.vg1x2.nxv8f16(i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 8 x half>, <vscale x 8 x half> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_za16_f16_vg1x2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sme.read.vg1x2.nxv8f16(i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 8 x half>, <vscale x 8 x half> } [[TMP0]]
//
svfloat16x2_t test_svread_za16_f16_vg1x2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_za16_f16_vg1x2(base);
}

// CHECK-LABEL: @test_svread_za32_s32_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.read.vg1x2.nxv4i32(i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_za32_s32_vg1x2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.read.vg1x2.nxv4i32(i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]]
//
svint32x2_t test_svread_za32_s32_vg1x2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_za32_s32_vg1x2(base);
}

// CHECK-LABEL: @test_svread_za32_u32_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.read.vg1x2.nxv4i32(i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_za32_u32_vg1x2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.read.vg1x2.nxv4i32(i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]]
//
svuint32x2_t test_svread_za32_u32_vg1x2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_za32_u32_vg1x2(base);
}

// CHECK-LABEL: @test_svread_za32_f32_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sme.read.vg1x2.nxv4f32(i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 4 x float>, <vscale x 4 x float> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_za32_f32_vg1x2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sme.read.vg1x2.nxv4f32(i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 4 x float>, <vscale x 4 x float> } [[TMP0]]
//
svfloat32x2_t test_svread_za32_f32_vg1x2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_za32_f32_vg1x2(base);
}

// CHECK-LABEL: @test_svread_za64_u64_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sme.read.vg1x2.nxv2i64(i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_za64_u64_vg1x2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sme.read.vg1x2.nxv2i64(i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64> } [[TMP0]]
//
svuint64x2_t test_svread_za64_u64_vg1x2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_za64_u64_vg1x2(base);
}

// CHECK-LABEL: @test_svread_za64_f64_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.aarch64.sme.read.vg1x2.nxv2f64(i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 2 x double>, <vscale x 2 x double> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_za64_f64_vg1x2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.aarch64.sme.read.vg1x2.nxv2f64(i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 2 x double>, <vscale x 2 x double> } [[TMP0]]
//
svfloat64x2_t test_svread_za64_f64_vg1x2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_za64_f64_vg1x2(base);
}

// CHECK-LABEL: @test_svread_za64_s64_vg1x2(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sme.read.vg1x2.nxv2i64(i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_za64_s64_vg1x2j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sme.read.vg1x2.nxv2i64(i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64> } [[TMP0]]
//
svint64x2_t test_svread_za64_s64_vg1x2(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_za64_s64_vg1x2(base);
}

// CHECK-LABEL: @test_svread_za8_s8_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.read.vg1x4.nxv16i8(i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z24test_svread_za8_s8_vg1x4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.read.vg1x4.nxv16i8(i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]]
//
svint8x4_t test_svread_za8_s8_vg1x4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_za8_s8_vg1x4(base);
}

// CHECK-LABEL: @test_svread_za8_u8_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.read.vg1x4.nxv16i8(i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z24test_svread_za8_u8_vg1x4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.aarch64.sme.read.vg1x4.nxv16i8(i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } [[TMP0]]
//
svuint8x4_t test_svread_za8_u8_vg1x4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_za8_u8_vg1x4(base);
}

// CHECK-LABEL: @test_svread_za16_s16_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.read.vg1x4.nxv8i16(i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_za16_s16_vg1x4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.read.vg1x4.nxv8i16(i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]]
//
svint16x4_t test_svread_za16_s16_vg1x4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_za16_s16_vg1x4(base);
}

// CHECK-LABEL: @test_svread_za16_u16_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.read.vg1x4.nxv8i16(i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_za16_u16_vg1x4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.aarch64.sme.read.vg1x4.nxv8i16(i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16>, <vscale x 8 x i16> } [[TMP0]]
//
svuint16x4_t test_svread_za16_u16_vg1x4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_za16_u16_vg1x4(base);
}

// CHECK-LABEL: @test_svread_za16_bf16_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.aarch64.sme.read.vg1x4.nxv8bf16(i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z27test_svread_za16_bf16_vg1x4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.aarch64.sme.read.vg1x4.nxv8bf16(i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
//
svbfloat16x4_t test_svread_za16_bf16_vg1x4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_za16_bf16_vg1x4(base);
}

// CHECK-LABEL: @test_svread_za16_f16_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sme.read.vg1x4.nxv8f16(i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_za16_f16_vg1x4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half> } @llvm.aarch64.sme.read.vg1x4.nxv8f16(i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half>, <vscale x 8 x half> } [[TMP0]]
//
svfloat16x4_t test_svread_za16_f16_vg1x4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_za16_f16_vg1x4(base);
}

// CHECK-LABEL: @test_svread_za32_s32_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.read.vg1x4.nxv4i32(i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_za32_s32_vg1x4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.read.vg1x4.nxv4i32(i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]]
//
svint32x4_t test_svread_za32_s32_vg1x4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_za32_s32_vg1x4(base);
}

// CHECK-LABEL: @test_svread_za32_u32_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.read.vg1x4.nxv4i32(i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_za32_u32_vg1x4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.aarch64.sme.read.vg1x4.nxv4i32(i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i32> } [[TMP0]]
//
svuint32x4_t test_svread_za32_u32_vg1x4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_za32_u32_vg1x4(base);
}

// CHECK-LABEL: @test_svread_za32_f32_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sme.read.vg1x4.nxv4f32(i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_za32_f32_vg1x4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float> } @llvm.aarch64.sme.read.vg1x4.nxv4f32(i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float>, <vscale x 4 x float> } [[TMP0]]
//
svfloat32x4_t test_svread_za32_f32_vg1x4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_za32_f32_vg1x4(base);
}

// CHECK-LABEL: @test_svread_za64_u64_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sme.read.vg1x4.nxv2i64(i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_za64_u64_vg1x4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sme.read.vg1x4.nxv2i64(i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } [[TMP0]]
//
svuint64x4_t test_svread_za64_u64_vg1x4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_za64_u64_vg1x4(base);
}

// CHECK-LABEL: @test_svread_za64_f64_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double> } @llvm.aarch64.sme.read.vg1x4.nxv2f64(i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_za64_f64_vg1x4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double> } @llvm.aarch64.sme.read.vg1x4.nxv2f64(i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double>, <vscale x 2 x double> } [[TMP0]]
//
svfloat64x4_t test_svread_za64_f64_vg1x4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_za64_f64_vg1x4(base);
}

// CHECK-LABEL: @test_svread_za64_s64_vg1x4(
// CHECK-NEXT:  entry:
// CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sme.read.vg1x4.nxv2i64(i32 [[BASE:%.*]])
// CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } [[TMP0]]
//
// CPP-CHECK-LABEL: @_Z26test_svread_za64_s64_vg1x4j(
// CPP-CHECK-NEXT:  entry:
// CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.aarch64.sme.read.vg1x4.nxv2i64(i32 [[BASE:%.*]])
// CPP-CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64> } [[TMP0]]
//
svint64x4_t test_svread_za64_s64_vg1x4(uint32_t base) __arm_streaming __arm_in("za") {
  return svread_za64_s64_vg1x4(base);
}