// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 5

// REQUIRES: aarch64-registered-target

// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sme -target-feature +sme2 -target-feature +sme-lutv2 -disable-O0-optnone -Werror -Wall -emit-llvm -o - %s | opt -S -p mem2reg,instcombine,tailcallelim | FileCheck %s
// RUN: %clang_cc1 -triple aarch64-none-linux-gnu -target-feature +sme -target-feature +sme2 -target-feature +sme-lutv2 -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-none-linux-gnu -target-feature +sme -target-feature +sme2 -target-feature +sme-lutv2 -disable-O0-optnone -Werror -Wall -o /dev/null %s


#include <arm_sme.h>

// CHECK-LABEL: define dso_local { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @test_luti4_zt_u8_x4(
// CHECK-SAME: <vscale x 16 x i8> [[OP_COERCE0:%.*]], <vscale x 16 x i8> [[OP_COERCE1:%.*]]) #[[ATTR0:[0-9]+]] {
// 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.luti4.zt.x4.nxv16i8(i32 0, <vscale x 16 x i8> [[OP_COERCE0]], <vscale x 16 x i8> [[OP_COERCE1]])
// 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: define dso_local { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @_Z19test_luti4_zt_u8_x411svuint8x2_t(
// CPP-CHECK-SAME: <vscale x 16 x i8> [[OP_COERCE0:%.*]], <vscale x 16 x i8> [[OP_COERCE1:%.*]]) #[[ATTR0:[0-9]+]] {
// 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.luti4.zt.x4.nxv16i8(i32 0, <vscale x 16 x i8> [[OP_COERCE0]], <vscale x 16 x i8> [[OP_COERCE1]])
// 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_luti4_zt_u8_x4(svuint8x2_t op)  __arm_streaming __arm_in("zt0") {
  return svluti4_zt_u8_x4(0, op);
}

// CHECK-LABEL: define dso_local { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @test_luti4_zt_s8_x4(
// CHECK-SAME: <vscale x 16 x i8> [[OP_COERCE0:%.*]], <vscale x 16 x i8> [[OP_COERCE1:%.*]]) #[[ATTR0]] {
// 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.luti4.zt.x4.nxv16i8(i32 0, <vscale x 16 x i8> [[OP_COERCE0]], <vscale x 16 x i8> [[OP_COERCE1]])
// 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: define dso_local { <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8>, <vscale x 16 x i8> } @_Z19test_luti4_zt_s8_x411svuint8x2_t(
// CPP-CHECK-SAME: <vscale x 16 x i8> [[OP_COERCE0:%.*]], <vscale x 16 x i8> [[OP_COERCE1:%.*]]) #[[ATTR0]] {
// 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.luti4.zt.x4.nxv16i8(i32 0, <vscale x 16 x i8> [[OP_COERCE0]], <vscale x 16 x i8> [[OP_COERCE1]])
// 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_luti4_zt_s8_x4(svuint8x2_t op)  __arm_streaming __arm_in("zt0") {
  return svluti4_zt_s8_x4(0, op);
}