[WebAssembly] Implementation of intrinsic for ref.null and HeapType removal

[pmatos]

Summary

This patch implements the intrinsic for ref.null.
In the process of implementing int_wasm_ref_null_func() and
int_wasm_ref_null_extern() intrinsics, it removes the redundant
HeapType.

This also causes the textual assembler syntax for ref.null to
change. Instead of receiving an argument: func or extern, the
instruction mnemonic is either ref.null_func or ref.null_extern,
without the need for a further operand.

[tlively]

LGTM modulo comments. It's nice that we get to delete so much code this way.

llvm/lib/IR/Function.cpp

930–944

It would be good to undo these unrelated formatting changes.

llvm/lib/Target/WebAssembly/MCTargetDesc/WebAssemblyMCCodeEmitter.cpp

91

Can we roll these values into the opcodes defined in the tablegen files? That would remove the need for this exception and would make the disassembler recover the correct instruction. The only issue I can think of is that there might be code paths that depend on the opcodes being in a certain range that this would exceed.

[pmatos]

llvm/lib/IR/Function.cpp

930–944

Yes, well, I want clang-format and it suggested the change, which is why I accepted it but fair enough... I will roll them back.

llvm/lib/Target/WebAssembly/MCTargetDesc/WebAssemblyMCCodeEmitter.cpp

91

Ah - love the suggestion. Completely forgot I could and should roll these into tablegen. Also bonus points as I learned to use !cond(...).

llvm/include/llvm/IR/Intrinsics.td

 def llvm_v1f64_ty      : LLVMType<v1f64>;    //  1 x double
 def llvm_v2f64_ty      : LLVMType<v2f64>;    //  2 x double
 def llvm_v4f64_ty      : LLVMType<v4f64>;    //  4 x double
 def llvm_v8f64_ty      : LLVMType<v8f64>;    //  8 x double
 def llvm_v16f64_ty     : LLVMType<v16f64>;   // 16 x double
 
 def llvm_vararg_ty     : LLVMType<isVoid>;   // this means vararg here
 
+def llvm_externref_ty  : LLVMType<externref>;
+def llvm_funcref_ty    : LLVMType<funcref>;
+
 //===----------------------------------------------------------------------===//
 // Intrinsic Definitions.
 //===----------------------------------------------------------------------===//
 
 // Intrinsic class - This is used to define one LLVM intrinsic.  The name of the
 // intrinsic definition should start with "int_", then match the LLVM intrinsic
 // name with the "llvm." prefix removed, and all "."s turned into "_"s.  For
 // example, llvm.bswap.i16 -> int_bswap_i16.

llvm/include/llvm/IR/IntrinsicsWebAssembly.td

 def int_wasm_memory_size : Intrinsic<[llvm_anyint_ty],
                                      [llvm_i32_ty],
                                      [IntrReadMem]>;
 def int_wasm_memory_grow : Intrinsic<[llvm_anyint_ty],
                                      [llvm_i32_ty, LLVMMatchType<0>],
                                      []>;
 
 //===----------------------------------------------------------------------===//
+// ref.null intrinsics
+//===----------------------------------------------------------------------===//
+def int_wasm_ref_null_extern : Intrinsic<[llvm_externref_ty], [], [IntrNoMem]>;
+def int_wasm_ref_null_func : Intrinsic<[llvm_funcref_ty], [], [IntrNoMem]>;
+
+//===----------------------------------------------------------------------===//
 // Trapping float-to-int conversions
 //===----------------------------------------------------------------------===//
 
 def int_wasm_trunc_signed : Intrinsic<[llvm_anyint_ty],
                                       [llvm_anyfloat_ty],
                                       [IntrNoMem]>;
 def int_wasm_trunc_unsigned : Intrinsic<[llvm_anyint_ty],
                                         [llvm_anyfloat_ty],

llvm/lib/IR/Function.cpp

 
 /// IIT_Info - These are enumerators that describe the entries returned by the
 /// getIntrinsicInfoTableEntries function.
 ///
 /// NOTE: This must be kept in synch with the copy in TblGen/IntrinsicEmitter!
 enum IIT_Info {
   // Common values should be encoded with 0-15.
   IIT_Done = 0,
-  IIT_I1   = 1,
-  IIT_I8   = 2,
-  IIT_I16  = 3,
-  IIT_I32  = 4,
-  IIT_I64  = 5,
-  IIT_F16  = 6,
-  IIT_F32  = 7,
-  IIT_F64  = 8,
-  IIT_V2   = 9,
-  IIT_V4   = 10,
-  IIT_V8   = 11,
-  IIT_V16  = 12,
-  IIT_V32  = 13,
-  IIT_PTR  = 14,
-  IIT_ARG  = 15,
+  IIT_I1 = 1,
+  IIT_I8 = 2,
+  IIT_I16 = 3,
+  IIT_I32 = 4,
+  IIT_I64 = 5,
+  IIT_F16 = 6,
+  IIT_F32 = 7,
+  IIT_F64 = 8,
+  IIT_V2 = 9,
+  IIT_V4 = 10,
+  IIT_V8 = 11,
+  IIT_V16 = 12,
+  IIT_V32 = 13,
+  IIT_PTR = 14,
+  IIT_ARG = 15,
 
   // Values from 16+ are only encodable with the inefficient encoding.
-  IIT_V64  = 16,
-  IIT_MMX  = 17,
+  IIT_V64 = 16,
+  IIT_MMX = 17,
   IIT_TOKEN = 18,
   IIT_METADATA = 19,
   IIT_EMPTYSTRUCT = 20,
   IIT_STRUCT2 = 21,
   IIT_STRUCT3 = 22,
   IIT_STRUCT4 = 23,
   IIT_STRUCT5 = 24,
   IIT_EXTEND_ARG = 25,
   IIT_TRUNC_ARG = 26,
   IIT_ANYPTR = 27,
-  IIT_V1   = 28,
+  IIT_V1 = 28,
   IIT_VARARG = 29,
   IIT_HALF_VEC_ARG = 30,
   IIT_SAME_VEC_WIDTH_ARG = 31,
   IIT_PTR_TO_ARG = 32,
   IIT_PTR_TO_ELT = 33,
   IIT_VEC_OF_ANYPTRS_TO_ELT = 34,
   IIT_I128 = 35,
   IIT_V512 = 36,
   IIT_V1024 = 37,
   IIT_STRUCT6 = 38,
   IIT_STRUCT7 = 39,
   IIT_STRUCT8 = 40,
   IIT_F128 = 41,
   IIT_VEC_ELEMENT = 42,
   IIT_SCALABLE_VEC = 43,
   IIT_SUBDIVIDE2_ARG = 44,
   IIT_SUBDIVIDE4_ARG = 45,
   IIT_VEC_OF_BITCASTS_TO_INT = 46,
   IIT_V128 = 47,
   IIT_BF16 = 48,
   IIT_STRUCT9 = 49,
   IIT_V256 = 50,
-  IIT_AMX  = 51,
-  IIT_PPCF128 = 52
+  IIT_AMX = 51,
+  IIT_PPCF128 = 52,
+  IIT_EXTERNREF = 53,
+  IIT_FUNCREF = 54
 };
 
 static void DecodeIITType(unsigned &NextElt, ArrayRef<unsigned char> Infos,
                       IIT_Info LastInfo,
                       SmallVectorImpl<Intrinsic::IITDescriptor> &OutputTable) {
   using namespace Intrinsic;
 
   bool IsScalableVector = (LastInfo == IIT_SCALABLE_VEC);
static void DecodeIITType(unsigned &NextElt, ArrayRef<unsigned char> Infos,
   case IIT_V512:
     OutputTable.push_back(IITDescriptor::getVector(512, IsScalableVector));
     DecodeIITType(NextElt, Infos, Info, OutputTable);
     return;
   case IIT_V1024:
     OutputTable.push_back(IITDescriptor::getVector(1024, IsScalableVector));
     DecodeIITType(NextElt, Infos, Info, OutputTable);
     return;
+  case IIT_EXTERNREF:
+    OutputTable.push_back(IITDescriptor::get(IITDescriptor::Pointer, 10));
+    OutputTable.push_back(IITDescriptor::get(IITDescriptor::Struct, 0));
+    return;
+  case IIT_FUNCREF:
+    OutputTable.push_back(IITDescriptor::get(IITDescriptor::Pointer, 20));
+    OutputTable.push_back(IITDescriptor::get(IITDescriptor::Integer, 8));
+    return;
   case IIT_PTR:
     OutputTable.push_back(IITDescriptor::get(IITDescriptor::Pointer, 0));
     DecodeIITType(NextElt, Infos, Info, OutputTable);
     return;
   case IIT_ANYPTR: {  // [ANYPTR addrspace, subtype]
     OutputTable.push_back(IITDescriptor::get(IITDescriptor::Pointer,
                                              Infos[NextElt++]));
     DecodeIITType(NextElt, Infos, Info, OutputTable);

llvm/lib/Target/WebAssembly/AsmParser/WebAssemblyAsmParser.cpp

bool ParseInstruction(ParseInstructionInfo & /*Info*/, StringRef Name,
     Operands.push_back(std::make_unique<WebAssemblyOperand>(
         WebAssemblyOperand::Token, NameLoc, SMLoc::getFromPointer(Name.end()),
         WebAssemblyOperand::TokOp{Name}));
 
     // If this instruction is part of a control flow structure, ensure
     // proper nesting.
     bool ExpectBlockType = false;
     bool ExpectFuncType = false;
-    bool ExpectHeapType = false;
     std::unique_ptr<WebAssemblyOperand> FunctionTable;
     if (Name == "block") {
       push(Block);
       ExpectBlockType = true;
     } else if (Name == "loop") {
       push(Loop);
       ExpectBlockType = true;
     } else if (Name == "try") {
if (Name == "block") {
         return true;
     } else if (Name == "call_indirect" || Name == "return_call_indirect") {
       // These instructions have differing operand orders in the text format vs
       // the binary formats.  The MC instructions follow the binary format, so
       // here we stash away the operand and append it later.
       if (parseFunctionTableOperand(&FunctionTable))
         return true;
       ExpectFuncType = true;
-    } else if (Name == "ref.null") {
-      ExpectHeapType = true;
     }
 
     if (ExpectFuncType || (ExpectBlockType && Lexer.is(AsmToken::LParen))) {
       // This has a special TYPEINDEX operand which in text we
       // represent as a signature, such that we can re-build this signature,
       // attach it to an anonymous symbol, which is what WasmObjectWriter
       // expects to be able to recreate the actual unique-ified type indices.
       auto Loc = Parser.getTok();
while (Lexer.isNot(AsmToken::EndOfStatement)) {
         auto &Id = Lexer.getTok();
         if (ExpectBlockType) {
           // Assume this identifier is a block_type.
           auto BT = WebAssembly::parseBlockType(Id.getString());
           if (BT == WebAssembly::BlockType::Invalid)
             return error("Unknown block type: ", Id);
           addBlockTypeOperand(Operands, NameLoc, BT);
           Parser.Lex();
-        } else if (ExpectHeapType) {
-          auto HeapType = WebAssembly::parseHeapType(Id.getString());
-          if (HeapType == WebAssembly::HeapType::Invalid) {
-            return error("Expected a heap type: ", Id);
-          }
-          Operands.push_back(std::make_unique<WebAssemblyOperand>(
-              WebAssemblyOperand::Integer, Id.getLoc(), Id.getEndLoc(),
-              WebAssemblyOperand::IntOp{static_cast<int64_t>(HeapType)}));
-          Parser.Lex();
         } else {
           // Assume this identifier is a label.
           const MCExpr *Val;
           SMLoc End;
           if (Parser.parseExpression(Val, End))
             return error("Cannot parse symbol: ", Lexer.getTok());
           Operands.push_back(std::make_unique<WebAssemblyOperand>(
               WebAssemblyOperand::Symbol, Id.getLoc(), Id.getEndLoc(),

llvm/lib/Target/WebAssembly/Disassembler/WebAssemblyDisassembler.cpp

case WebAssembly::OPERAND_SIGNATURE: {
         auto *WasmSym = cast<MCSymbolWasm>(Sym);
         WasmSym->setType(wasm::WASM_SYMBOL_TYPE_FUNCTION);
         const MCExpr *Expr = MCSymbolRefExpr::create(
             WasmSym, MCSymbolRefExpr::VK_WASM_TYPEINDEX, getContext());
         MI.addOperand(MCOperand::createExpr(Expr));
       }
       break;
     }
-    // heap_type operands, for e.g. ref.null:
-    case WebAssembly::OPERAND_HEAPTYPE: {
-      int64_t Val;
-      uint64_t PrevSize = Size;
-      if (!nextLEB(Val, Bytes, Size, true))
-        return MCDisassembler::Fail;
-      if (Val < 0 && Size == PrevSize + 1) {
-        // The HeapType encoding is like BlockType, in that encodings that
-        // decode as negative values indicate ValTypes.  In practice we expect
-        // either wasm::ValType::EXTERNREF or wasm::ValType::FUNCREF here.
-        //
-        // The positive SLEB values are reserved for future expansion and are
-        // expected to be type indices in the typed function references
-        // proposal, and should disassemble as MCSymbolRefExpr as in BlockType
-        // above.
-        MI.addOperand(MCOperand::createImm(Val & 0x7f));
-      } else {
-        MI.addOperand(
-            MCOperand::createImm(int64_t(WebAssembly::HeapType::Invalid)));
-      }
-      break;
-    }
     // FP operands.
     case WebAssembly::OPERAND_F32IMM: {
       if (!parseImmediate<float>(MI, Size, Bytes))
         return MCDisassembler::Fail;
       break;
     }
     case WebAssembly::OPERAND_F64IMM: {
       if (!parseImmediate<double>(MI, Size, Bytes))

llvm/lib/Target/WebAssembly/MCTargetDesc/WebAssemblyInstPrinter.h

public:
   // Used by tblegen code.
   void printOperand(const MCInst *MI, unsigned OpNo, raw_ostream &O,
                     bool IsVariadicDef = false);
   void printBrList(const MCInst *MI, unsigned OpNo, raw_ostream &O);
   void printWebAssemblyP2AlignOperand(const MCInst *MI, unsigned OpNo,
                                       raw_ostream &O);
   void printWebAssemblySignatureOperand(const MCInst *MI, unsigned OpNo,
                                         raw_ostream &O);
-  void printWebAssemblyHeapTypeOperand(const MCInst *MI, unsigned OpNo,
-                                       raw_ostream &O);
 
   // Autogenerated by tblgen.
   std::pair<const char *, uint64_t> getMnemonic(const MCInst *MI) override;
   void printInstruction(const MCInst *MI, uint64_t Address, raw_ostream &O);
   static const char *getRegisterName(unsigned RegNo);
 };
 
 } // end namespace llvm
 
 #endif

llvm/lib/Target/WebAssembly/MCTargetDesc/WebAssemblyInstPrinter.cpp

if (Op.isImm()) {
     if (Sym->getSignature()) {
       O << WebAssembly::signatureToString(Sym->getSignature());
     } else {
       // Disassembler does not currently produce a signature
       O << "unknown_type";
     }
   }
 }
-
-void WebAssemblyInstPrinter::printWebAssemblyHeapTypeOperand(const MCInst *MI,
-                                                             unsigned OpNo,
-                                                             raw_ostream &O) {
-  const MCOperand &Op = MI->getOperand(OpNo);
-  if (Op.isImm()) {
-    switch (Op.getImm()) {
-    case long(wasm::ValType::EXTERNREF):
-      O << "extern";
-      break;
-    case long(wasm::ValType::FUNCREF):
-      O << "func";
-      break;
-    default:
-      O << "unsupported_heap_type_value";
-      break;
-    }
-  } else {
-    // Typed function references and other subtypes of funcref and externref
-    // currently unimplemented.
-    O << "unsupported_heap_type_operand";
-  }
-}

llvm/lib/Target/WebAssembly/MCTargetDesc/WebAssemblyMCCodeEmitter.cpp

void WebAssemblyMCCodeEmitter::encodeInstruction(
   // each table entry, and the default operand.
   if (MI.getOpcode() == WebAssembly::BR_TABLE_I32_S ||
       MI.getOpcode() == WebAssembly::BR_TABLE_I64_S)
     encodeULEB128(MI.getNumOperands() - 1, OS);
   if (MI.getOpcode() == WebAssembly::BR_TABLE_I32 ||
       MI.getOpcode() == WebAssembly::BR_TABLE_I64)
     encodeULEB128(MI.getNumOperands() - 2, OS);
 
+  // For REF_NULL, which has no HeapType explicit argument we just explicitly
+  // emit the encoding for the operand manually
+  if (MI.getOpcode() == WebAssembly::REF_NULL_FUNCREF ||
+      MI.getOpcode() == WebAssembly::REF_NULL_FUNCREF_S)
+    OS << uint8_t(0x70);
+  if (MI.getOpcode() == WebAssembly::REF_NULL_EXTERNREF ||
+      MI.getOpcode() == WebAssembly::REF_NULL_EXTERNREF_S)
+    OS << uint8_t(0x6f);
+
   const MCInstrDesc &Desc = MCII.get(MI.getOpcode());
   for (unsigned I = 0, E = MI.getNumOperands(); I < E; ++I) {
     const MCOperand &MO = MI.getOperand(I);
     if (MO.isReg()) {
       /* nothing to encode */
 
     } else if (MO.isImm()) {
       if (I < Desc.getNumOperands()) {
         const MCOperandInfo &Info = Desc.OpInfo[I];
         LLVM_DEBUG(dbgs() << "Encoding immediate: type="
                           << int(Info.OperandType) << "\n");
         switch (Info.OperandType) {
         case WebAssembly::OPERAND_I32IMM:
           encodeSLEB128(int32_t(MO.getImm()), OS);
           break;
         case WebAssembly::OPERAND_OFFSET32:
           encodeULEB128(uint32_t(MO.getImm()), OS);
           break;
         case WebAssembly::OPERAND_I64IMM:
           encodeSLEB128(int64_t(MO.getImm()), OS);
           break;
         case WebAssembly::OPERAND_SIGNATURE:
-        case WebAssembly::OPERAND_HEAPTYPE:
-          OS << uint8_t(MO.getImm());
-          break;
         case WebAssembly::OPERAND_VEC_I8IMM:
           support::endian::write<uint8_t>(OS, MO.getImm(), support::little);
           break;
         case WebAssembly::OPERAND_VEC_I16IMM:
           support::endian::write<uint16_t>(OS, MO.getImm(), support::little);
           break;
         case WebAssembly::OPERAND_VEC_I32IMM:
           support::endian::write<uint32_t>(OS, MO.getImm(), support::little);

llvm/lib/Target/WebAssembly/MCTargetDesc/WebAssemblyMCTargetDesc.h

enum OperandType {
   /// type signature immediate for call_indirect.
   OPERAND_TYPEINDEX,
   /// Tag index.
   OPERAND_TAG,
   /// A list of branch targets for br_list.
   OPERAND_BRLIST,
   /// 32-bit unsigned table number.
   OPERAND_TABLE,
-  /// heap type immediate for ref.null.
-  OPERAND_HEAPTYPE,
 };
 } // end namespace WebAssembly
 
 namespace WebAssemblyII {
 
 /// Target Operand Flag enum.
 enum TOF {
   MO_NO_FLAG = 0,

llvm/lib/Target/WebAssembly/Utils/WebAssemblyTypeUtilities.h

enum class BlockType : unsigned {
   // Multivalue blocks (and other non-void blocks) are only emitted when the
   // blocks will never be exited and are at the ends of functions (see
   // WebAssemblyCFGStackify::fixEndsAtEndOfFunction). They also are never made
   // to pop values off the stack, so the exact multivalue signature can always
   // be inferred from the return type of the parent function in MCInstLower.
   Multivalue = 0xffff,
 };
 
-/// Used as immediate MachineOperands for heap types, e.g. for ref.null.
-enum class HeapType : unsigned {
-  Invalid = 0x00,
-  Externref = unsigned(wasm::ValType::EXTERNREF),
-  Funcref = unsigned(wasm::ValType::FUNCREF),
-};
-
 // Convert StringRef to ValType / HealType / BlockType
 
 Optional<wasm::ValType> parseType(StringRef Type);
-HeapType parseHeapType(StringRef Type);
 BlockType parseBlockType(StringRef Type);
 MVT parseMVT(StringRef Type);
 
 // Convert ValType or a list/signature of ValTypes to a string.
 
 // Convert an unsinged integer, which can be among wasm::ValType enum, to its
 // type name string. If the input is not within wasm::ValType, returns
 // "invalid_type".

llvm/lib/Target/WebAssembly/Utils/WebAssemblyTypeUtilities.cpp

if (Type == "v128" || Type == "i8x16" || Type == "i16x8" || Type == "i32x4" ||
     return wasm::ValType::V128;
   if (Type == "funcref")
     return wasm::ValType::FUNCREF;
   if (Type == "externref")
     return wasm::ValType::EXTERNREF;
   return Optional<wasm::ValType>();
 }
 
-WebAssembly::HeapType WebAssembly::parseHeapType(StringRef Type) {
-  return StringSwitch<WebAssembly::HeapType>(Type)
-      .Case("extern", WebAssembly::HeapType::Externref)
-      .Case("func", WebAssembly::HeapType::Funcref)
-      .Default(WebAssembly::HeapType::Invalid);
-}
-
 WebAssembly::BlockType WebAssembly::parseBlockType(StringRef Type) {
   // Multivalue block types are handled separately in parseSignature
   return StringSwitch<WebAssembly::BlockType>(Type)
       .Case("i32", WebAssembly::BlockType::I32)
       .Case("i64", WebAssembly::BlockType::I64)
       .Case("f32", WebAssembly::BlockType::F32)
       .Case("f64", WebAssembly::BlockType::F64)
       .Case("v128", WebAssembly::BlockType::V128)

llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp

Register RegZero =
         MF.getRegInfo().createVirtualRegister(&WebAssembly::I32RegClass);
     MachineInstr *Const0 =
         BuildMI(MF, DL, TII.get(WebAssembly::CONST_I32), RegZero).addImm(0);
     BB->insertAfter(MIB.getInstr()->getIterator(), Const0);
 
     Register RegFuncref =
         MF.getRegInfo().createVirtualRegister(&WebAssembly::FUNCREFRegClass);
     MachineInstr *RefNull =
-        BuildMI(MF, DL, TII.get(WebAssembly::REF_NULL_FUNCREF), RegFuncref)
-            .addImm(static_cast<int32_t>(WebAssembly::HeapType::Funcref));
+        BuildMI(MF, DL, TII.get(WebAssembly::REF_NULL_FUNCREF), RegFuncref);
     BB->insertAfter(Const0->getIterator(), RefNull);
 
     MachineInstr *TableSet =
         BuildMI(MF, DL, TII.get(WebAssembly::TABLE_SET_FUNCREF))
             .addSym(Table)
             .addReg(RegZero)
             .addReg(RegFuncref);
     BB->insertAfter(RefNull->getIterator(), TableSet);

llvm/lib/Target/WebAssembly/WebAssemblyInstrInfo.td

 
 } // OperandType = "OPERAND_P2ALIGN"
 
 let OperandType = "OPERAND_SIGNATURE" in
 def Signature : Operand<i32> {
   let PrintMethod = "printWebAssemblySignatureOperand";
 }
 
-let OperandType = "OPERAND_HEAPTYPE" in
-def HeapType : Operand<i32> {
-  let PrintMethod = "printWebAssemblyHeapTypeOperand";
-}
-
 let OperandType = "OPERAND_TYPEINDEX" in
 def TypeIndex : Operand<i32>;
 
 } // OperandNamespace = "WebAssembly"
 
 // TODO: Find more places to use this.
 def bool_node : PatLeaf<(i32 I32:$cond), [{
   return CurDAG->computeKnownBits(SDValue(N, 0)).countMinLeadingZeros() == 31;

llvm/lib/Target/WebAssembly/WebAssemblyInstrRef.td

 // WebAssemblyInstrRef.td - WebAssembly reference type codegen --*- tablegen -*-
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 ///
 /// \file
 /// WebAssembly reference type operand codegen constructs.
 ///
 //===----------------------------------------------------------------------===//
 
-multiclass REF_I<WebAssemblyRegClass rc, ValueType vt> {
-  defm REF_NULL_#rc : I<(outs rc:$res), (ins HeapType:$heaptype),
-                        (outs), (ins HeapType:$heaptype),
-                        [],
-                        "ref.null\t$res, $heaptype",
-                        "ref.null\t$heaptype",
+multiclass REF_I<WebAssemblyRegClass rc, ValueType vt, string ht> {
+  defm REF_NULL_#rc : I<(outs rc:$dst), (ins),
+                        (outs), (ins),
+                        [(set rc:$dst, (!cast<Intrinsic>("int_wasm_ref_null_" # ht)))],
+                        "ref.null_" # ht # "$dst",
+                        "ref.null_" # ht,
                         0xd0>,
                       Requires<[HasReferenceTypes]>;
   defm SELECT_#rc: I<(outs rc:$dst), (ins rc:$lhs, rc:$rhs, I32:$cond),
                      (outs), (ins),
                      [(set rc:$dst,
                        (select I32:$cond, rc:$lhs, rc:$rhs))],
                      vt#".select\t$dst, $lhs, $rhs, $cond",
                      vt#".select", 0x1b>,
                    Requires<[HasReferenceTypes]>;
 }
 
-defm "" : REF_I<FUNCREF, funcref>;
-defm "" : REF_I<EXTERNREF, externref>;
+defm "" : REF_I<FUNCREF, funcref, "func">;
+defm "" : REF_I<EXTERNREF, externref, "extern">;
 
 foreach rc = [FUNCREF, EXTERNREF] in {
 def : Pat<(select (i32 (setne I32:$cond, 0)), rc:$lhs, rc:$rhs),
           (!cast<Instruction>("SELECT_"#rc) rc:$lhs, rc:$rhs, I32:$cond)>;
 def : Pat<(select (i32 (seteq I32:$cond, 0)), rc:$lhs, rc:$rhs),
           (!cast<Instruction>("SELECT_"#rc) rc:$rhs, rc:$lhs, I32:$cond)>;
 }

llvm/lib/Target/WebAssembly/WebAssemblyMCInstLower.cpp

case MachineOperand::MO_Immediate: {
           assert(BT != WebAssembly::BlockType::Invalid);
           if (BT == WebAssembly::BlockType::Multivalue) {
             SmallVector<wasm::ValType, 1> Returns;
             getFunctionReturns(MI, Returns);
             MCOp = lowerTypeIndexOperand(std::move(Returns),
                                          SmallVector<wasm::ValType, 4>());
             break;
           }
-        } else if (Info.OperandType == WebAssembly::OPERAND_HEAPTYPE) {
-          assert(static_cast<WebAssembly::HeapType>(MO.getImm()) !=
-                 WebAssembly::HeapType::Invalid);
-          // With typed function references, this will need a case for type
-          // index operands.  Otherwise, fall through.
         }
       }
       MCOp = MCOperand::createImm(MO.getImm());
       break;
     }
     case MachineOperand::MO_FPImmediate: {
       const ConstantFP *Imm = MO.getFPImm();
       const uint64_t BitPattern =

llvm/test/CodeGen/WebAssembly/funcref-call.ll

 ; CHECK-LABEL: call_funcref:
 ; CHECK-NEXT: functype       call_funcref (funcref) -> ()
 ; CHECK-NEXT: i32.const 0
 ; CHECK-NEXT: local.get 0
 ; CHECK-NEXT: table.set __funcref_call_table
 ; CHECK-NEXT: i32.const 0
 ; CHECK-NEXT: call_indirect __funcref_call_table, () -> ()
 ; CHECK-NEXT: i32.const 0
-; CHECK-NEXT: ref.null func
+; CHECK-NEXT: ref.null_func
 ; CHECK-NEXT: table.set __funcref_call_table
 ; CHECK-NEXT: end_function
 

llvm/test/CodeGen/WebAssembly/funcref-table_call.ll

 ; CHECK-NEXT: .functype       call_funcref_from_table (i32) -> ()
 ; CHECK-NEXT: i32.const       0
 ; CHECK-NEXT: local.get       0
 ; CHECK-NEXT: table.get       funcref_table
 ; CHECK-NEXT: table.set       __funcref_call_table
 ; CHECK-NEXT: i32.const       0
 ; CHECK-NEXT: call_indirect    __funcref_call_table, () -> ()
 ; CHECK-NEXT: i32.const       0
-; CHECK-NEXT: ref.null        func
+; CHECK-NEXT: ref.null_func
 ; CHECK-NEXT: table.set       __funcref_call_table
 ; CHECK-NEXT: end_function
 
 
 ; CHECK:     .globl funcref_table
 ; CHECK-NEXT .globaltype     funcref_table, funcref

llvm/test/CodeGen/WebAssembly/ref-null.ll

+; RUN: llc --mtriple=wasm32-unknown-unknown -asm-verbose=false -mattr=+reference-types < %s | FileCheck %s
+
+%extern = type opaque
+%externref = type %extern addrspace(10)* ;; addrspace 10 is nonintegral
+%funcref = type i8 addrspace(20)*      ;; addrspace 20 is nonintegral
+
+declare %externref @llvm.wasm.ref.null.extern() nounwind
+declare %funcref @llvm.wasm.ref.null.func() nounwind
+
+define %externref @get_null_extern() {
+; CHECK-LABEL: get_null_extern:
+; CHECK-NEXT:  .functype       get_null_extern () -> (externref)
+; CHECK-NEXT:  ref.null_extern
+; CHECK-NEXT:  end_function
+  %null = call %externref @llvm.wasm.ref.null.extern()
+  ret %externref %null
+}
+
+define %funcref @get_null_func() {
+; CHECK-LABEL: get_null_func:
+; CHECK-NEXT:  .functype       get_null_func () -> (funcref)
+; CHECK-NEXT:  ref.null_func
+; CHECK-NEXT:  end_function
+  %null = call %funcref @llvm.wasm.ref.null.func()
+  ret %funcref %null
+}

llvm/test/MC/WebAssembly/reference-types.s

 # RUN: llvm-mc -show-encoding -triple=wasm32-unknown-unknown -mattr=+reference-types < %s | FileCheck %s
 # RUN: llvm-mc -show-encoding -triple=wasm64-unknown-unknown -mattr=+reference-types < %s | FileCheck %s
 
 # CHECK-LABEL: ref_null_test:
-# CHECK: ref.null func   # encoding: [0xd0,0x70]
-# CHECK: ref.null extern # encoding: [0xd0,0x6f]
+# CHECK: ref.null_func   # encoding: [0xd0,0x70]
+# CHECK: ref.null_extern # encoding: [0xd0,0x6f]
 ref_null_test:
   .functype ref_null_test () -> ()
-  ref.null func
+  ref.null_func
   drop
-  ref.null extern
+  ref.null_extern
   drop
   end_function
 
 # CHECK-LABEL: ref_sig_test_funcref:
 # CHECK-NEXT: .functype ref_sig_test_funcref (funcref) -> (funcref)
 ref_sig_test_funcref:
   .functype ref_sig_test_funcref (funcref) -> (funcref)
   local.get 0
   end_function
 
 # CHECK-LABEL: ref_sig_test_externref:
 # CHECK-NEXT: .functype ref_sig_test_externref (externref) -> (externref)
 ref_sig_test_externref:
   .functype ref_sig_test_externref (externref) -> (externref)
   local.get 0
   end_function
 
 # CHECK-LABEL: ref_select_test:
 # CHECK: funcref.select   # encoding: [0x1b]
 # CHECK: externref.select # encoding: [0x1b]
 ref_select_test:
   .functype ref_select_test () -> ()
-  ref.null func
-  ref.null func
+  ref.null_func
+  ref.null_func
   i32.const 0
   funcref.select
   drop
-  ref.null extern
-  ref.null extern
+  ref.null_extern
+  ref.null_extern
   i32.const 0
   externref.select
   drop
   end_function
 
 # CHECK-LABEL: ref_block_test:
 # CHECK: block funcref
 # CHECK: block externref
 ref_block_test:
   .functype ref_block_test () -> (externref, funcref)
   block funcref
   block externref
-  ref.null extern
+  ref.null_extern
   end_block
-  ref.null func
+  ref.null_func
   end_block
   end_function

llvm/utils/TableGen/IntrinsicEmitter.cpp

void IntrinsicEmitter::EmitIntrinsicToOverloadTable(
   OS << "#endif\n\n";
 }
 
 
 // NOTE: This must be kept in synch with the copy in lib/IR/Function.cpp!
 enum IIT_Info {
   // Common values should be encoded with 0-15.
   IIT_Done = 0,
-  IIT_I1   = 1,
-  IIT_I8   = 2,
-  IIT_I16  = 3,
-  IIT_I32  = 4,
-  IIT_I64  = 5,
-  IIT_F16  = 6,
-  IIT_F32  = 7,
-  IIT_F64  = 8,
-  IIT_V2   = 9,
-  IIT_V4   = 10,
-  IIT_V8   = 11,
-  IIT_V16  = 12,
-  IIT_V32  = 13,
-  IIT_PTR  = 14,
-  IIT_ARG  = 15,
+  IIT_I1 = 1,
+  IIT_I8 = 2,
+  IIT_I16 = 3,
+  IIT_I32 = 4,
+  IIT_I64 = 5,
+  IIT_F16 = 6,
+  IIT_F32 = 7,
+  IIT_F64 = 8,
+  IIT_V2 = 9,
+  IIT_V4 = 10,
+  IIT_V8 = 11,
+  IIT_V16 = 12,
+  IIT_V32 = 13,
+  IIT_PTR = 14,
+  IIT_ARG = 15,
 
   // Values from 16+ are only encodable with the inefficient encoding.
-  IIT_V64  = 16,
-  IIT_MMX  = 17,
+  IIT_V64 = 16,
+  IIT_MMX = 17,
   IIT_TOKEN = 18,
   IIT_METADATA = 19,
   IIT_EMPTYSTRUCT = 20,
   IIT_STRUCT2 = 21,
   IIT_STRUCT3 = 22,
   IIT_STRUCT4 = 23,
   IIT_STRUCT5 = 24,
   IIT_EXTEND_ARG = 25,
   IIT_TRUNC_ARG = 26,
   IIT_ANYPTR = 27,
-  IIT_V1   = 28,
+  IIT_V1 = 28,
   IIT_VARARG = 29,
   IIT_HALF_VEC_ARG = 30,
   IIT_SAME_VEC_WIDTH_ARG = 31,
   IIT_PTR_TO_ARG = 32,
   IIT_PTR_TO_ELT = 33,
   IIT_VEC_OF_ANYPTRS_TO_ELT = 34,
   IIT_I128 = 35,
   IIT_V512 = 36,
   IIT_V1024 = 37,
   IIT_STRUCT6 = 38,
   IIT_STRUCT7 = 39,
   IIT_STRUCT8 = 40,
   IIT_F128 = 41,
   IIT_VEC_ELEMENT = 42,
   IIT_SCALABLE_VEC = 43,
   IIT_SUBDIVIDE2_ARG = 44,
   IIT_SUBDIVIDE4_ARG = 45,
   IIT_VEC_OF_BITCASTS_TO_INT = 46,
   IIT_V128 = 47,
   IIT_BF16 = 48,
   IIT_STRUCT9 = 49,
   IIT_V256 = 50,
-  IIT_AMX  = 51,
-  IIT_PPCF128 = 52
+  IIT_AMX = 51,
+  IIT_PPCF128 = 52,
+  IIT_EXTERNREF = 53,
+  IIT_FUNCREF = 54
 };
 
 static void EncodeFixedValueType(MVT::SimpleValueType VT,
                                  std::vector<unsigned char> &Sig) {
   if (MVT(VT).isInteger()) {
     unsigned BitWidth = MVT(VT).getFixedSizeInBits();
     switch (BitWidth) {
     default: PrintFatalError("unhandled integer type width in intrinsic!");
static void EncodeFixedValueType(MVT::SimpleValueType VT,
   case MVT::token: return Sig.push_back(IIT_TOKEN);
   case MVT::Metadata: return Sig.push_back(IIT_METADATA);
   case MVT::x86mmx: return Sig.push_back(IIT_MMX);
   case MVT::x86amx: return Sig.push_back(IIT_AMX);
   // MVT::OtherVT is used to mean the empty struct type here.
   case MVT::Other: return Sig.push_back(IIT_EMPTYSTRUCT);
   // MVT::isVoid is used to represent varargs here.
   case MVT::isVoid: return Sig.push_back(IIT_VARARG);
+  case MVT::externref:
+    return Sig.push_back(IIT_EXTERNREF);
+  case MVT::funcref:
+    return Sig.push_back(IIT_FUNCREF);
   }
 }
 
 #if defined(_MSC_VER) && !defined(__clang__)
 #pragma optimize("",off) // MSVC 2015 optimizer can't deal with this function.
 #endif
 
 static void EncodeFixedType(Record *R, std::vector<unsigned char> &ArgCodes,