[IR] Introduce the ptrtoaddr instruction

llvm/docs/LangRef.rst

@@ -12521,6 +12521,59 @@ Example:
       %Y = ptrtoint ptr %P to i64                        ; yields zero extension on 32-bit architecture
       %Z = ptrtoint <4 x ptr> %P to <4 x i64>; yields vector zero extension for a vector of addresses on 32-bit architecture
 
+.. _i_ptrtoaddr:
+
+'``ptrtoaddr .. to``' Instruction
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Syntax:
+"""""""
+
+::
+
+      <result> = ptrtoaddr <ty> <value> to <ty2>             ; yields ty2
+
+Overview:
+"""""""""
+
+The '``ptrtoaddr``' instruction converts the pointer or a vector of
+pointers ``value`` to the underlying integer address (or vector of integers) of
+type ``ty2``. This is different from :ref:`ptrtoint <i_ptrtoint>` in that it
+only operates on the index bits of the pointer and ignores all other bits.
+``ty2`` must be the integer type (or vector of integers) matching the pointer
+index width of the address space of ``ty``.
+
+Arguments:
+""""""""""
+
+The '``ptrtoaddr``' instruction takes a ``value`` to cast, which must be
+a value of type :ref:`pointer <t_pointer>` or a vector of pointers, and a
+type to cast it to ``ty2``, which must be an :ref:`integer <t_integer>` or
+a vector of integers type.
+
+Semantics:
+""""""""""
+
+The '``ptrtoaddr``' instruction converts ``value`` to integer type ``ty2`` by
+interpreting the lowest index-width pointer representation bits as an integer.
+If the address size and the pointer representation size are the same and
+``value`` and ``ty2`` are the same size, then nothing is done (*no-op cast*)
+other than a type change.
+
+The ``ptrtoaddr`` instruction always :ref:`captures the address but not the provenance <pointercapture>`
+of the pointer argument.
+
+Example:
+""""""""
+This example assumes pointers in address space 1 are 64 bits in size with an
+address width of 32 bits (``p1:64:64:64:32`` :ref:`datalayout string<langref_datalayout>`)
+.. code-block:: llvm
+
+      %X = ptrtoaddr ptr addrspace(1) %P to i8  ; extracts low 32 bits and truncates
+      %Y = ptrtoaddr ptr addrspace(1) %P to i64 ; extracts low 32 bits and zero extends
+      %Z = ptrtoaddr <4 x ptr addrspace(1)> %P to <4 x i64>; yields vector zero extension of low 32 bits for each pointer
+
+
 .. _i_inttoptr:
 
 '``inttoptr .. to``' Instruction

llvm/docs/ReleaseNotes.md

@@ -56,6 +56,10 @@ Makes programs 10x faster by doing Special New Thing.
 Changes to the LLVM IR
 ----------------------
 
+* The `ptrtoaddr` instruction was introduced. This instruction returns the
+  address component of a pointer type variable but unlike `ptrtoint` does not
+  capture provenance ([#125687](https://github.com/llvm/llvm-project/pull/125687)).
+
 Changes to LLVM infrastructure
 ------------------------------
 

llvm/include/llvm-c/Core.h

@@ -111,6 +111,7 @@ typedef enum {
   LLVMFPTrunc        = 37,
   LLVMFPExt          = 38,
   LLVMPtrToInt       = 39,
+  LLVMPtrToAddr      = 69,
   LLVMIntToPtr       = 40,
   LLVMBitCast        = 41,
   LLVMAddrSpaceCast  = 60,

llvm/include/llvm/Analysis/IR2Vec.h

@@ -153,6 +153,7 @@ class Vocabulary {
                     static_cast<unsigned>(OperandKind::MaxOperandKind),
                 "OperandKindNames array size must match MaxOperandKind");
 
+public:
   /// Vocabulary layout constants
 #define LAST_OTHER_INST(NUM) static constexpr unsigned MaxOpcodes = NUM;
 #include "llvm/IR/Instruction.def"
@@ -162,7 +163,6 @@ class Vocabulary {
   static constexpr unsigned MaxOperandKinds =
       static_cast<unsigned>(OperandKind::MaxOperandKind);
 
-public:
   Vocabulary() = default;
   Vocabulary(VocabVector &&Vocab);
 

llvm/include/llvm/Analysis/TargetTransformInfoImpl.h

@@ -731,6 +731,12 @@ class TargetTransformInfoImplBase {
         return 0;
       break;
     }
+    case Instruction::PtrToAddr: {
+      unsigned DstSize = Dst->getScalarSizeInBits();
+      if (DL.isLegalInteger(DstSize) && DstSize >= DL.getAddressSizeInBits(Src))
+        return 0;
+      break;
+    }
     case Instruction::PtrToInt: {
       unsigned DstSize = Dst->getScalarSizeInBits();
       if (DL.isLegalInteger(DstSize) &&
@@ -1437,6 +1443,7 @@ class TargetTransformInfoImplCRTPBase : public TargetTransformInfoImplBase {
                                                Op2Info, Operands, I);
     }
     case Instruction::IntToPtr:
+    case Instruction::PtrToAddr:
     case Instruction::PtrToInt:
     case Instruction::SIToFP:
     case Instruction::UIToFP:

llvm/include/llvm/AsmParser/LLToken.h

@@ -319,6 +319,7 @@ enum Kind {
   kw_fptoui,
   kw_fptosi,
   kw_inttoptr,
+  kw_ptrtoaddr,
   kw_ptrtoint,
   kw_bitcast,
   kw_addrspacecast,

llvm/include/llvm/Bitcode/LLVMBitCodes.h

@@ -456,7 +456,8 @@ enum CastOpcodes {
   CAST_PTRTOINT = 9,
   CAST_INTTOPTR = 10,
   CAST_BITCAST = 11,
-  CAST_ADDRSPACECAST = 12
+  CAST_ADDRSPACECAST = 12,
+  CAST_PTRTOADDR = 13,
 };
 
 /// UnaryOpcodes - These are values used in the bitcode files to encode which

llvm/include/llvm/CodeGen/GlobalISel/IRTranslator.h

@@ -486,6 +486,10 @@ class IRTranslator : public MachineFunctionPass {
   bool translatePtrToInt(const User &U, MachineIRBuilder &MIRBuilder) {
     return translateCast(TargetOpcode::G_PTRTOINT, U, MIRBuilder);
   }
+  bool translatePtrToAddr(const User &U, MachineIRBuilder &MIRBuilder) {
+    // FIXME: this is not correct for pointers with addr width != pointer width
+    return translatePtrToInt(U, MIRBuilder);
+  }
   bool translateTrunc(const User &U, MachineIRBuilder &MIRBuilder) {
     return translateCast(TargetOpcode::G_TRUNC, U, MIRBuilder);
   }

llvm/include/llvm/IR/Constants.h

@@ -1158,6 +1158,8 @@ class ConstantExpr : public Constant {
   LLVM_ABI static Constant *getXor(Constant *C1, Constant *C2);
   LLVM_ABI static Constant *getTrunc(Constant *C, Type *Ty,
                                      bool OnlyIfReduced = false);
+  LLVM_ABI static Constant *getPtrToAddr(Constant *C, Type *Ty,
+                                         bool OnlyIfReduced = false);
   LLVM_ABI static Constant *getPtrToInt(Constant *C, Type *Ty,
                                         bool OnlyIfReduced = false);
   LLVM_ABI static Constant *getIntToPtr(Constant *C, Type *Ty,

llvm/include/llvm/IR/IRBuilder.h

@@ -2192,7 +2192,10 @@ class IRBuilderBase {
     return CreateCast(Instruction::FPExt, V, DestTy, Name, FPMathTag,
                       FMFSource);
   }
-
+  Value *CreatePtrToAddr(Value *V, const Twine &Name = "") {
+    return CreateCast(Instruction::PtrToInt, V,
+                      BB->getDataLayout().getAddressType(V->getType()), Name);
+  }
   Value *CreatePtrToInt(Value *V, Type *DestTy,
                         const Twine &Name = "") {
     return CreateCast(Instruction::PtrToInt, V, DestTy, Name);

llvm/include/llvm/IR/InstVisitor.h

@@ -183,6 +183,7 @@ class InstVisitor {
   RetTy visitUIToFPInst(UIToFPInst &I)            { DELEGATE(CastInst);}
   RetTy visitSIToFPInst(SIToFPInst &I)            { DELEGATE(CastInst);}
   RetTy visitPtrToIntInst(PtrToIntInst &I)        { DELEGATE(CastInst);}
+  RetTy visitPtrToAddrInst(PtrToAddrInst &I)      { DELEGATE(CastInst);}
   RetTy visitIntToPtrInst(IntToPtrInst &I)        { DELEGATE(CastInst);}
   RetTy visitBitCastInst(BitCastInst &I)          { DELEGATE(CastInst);}
   RetTy visitAddrSpaceCastInst(AddrSpaceCastInst &I) { DELEGATE(CastInst);}

llvm/include/llvm/IR/Instruction.def

@@ -190,35 +190,36 @@ HANDLE_CAST_INST(43, UIToFP  , UIToFPInst  )  // UInt -> floating point
 HANDLE_CAST_INST(44, SIToFP  , SIToFPInst  )  // SInt -> floating point
 HANDLE_CAST_INST(45, FPTrunc , FPTruncInst )  // Truncate floating point
 HANDLE_CAST_INST(46, FPExt   , FPExtInst   )  // Extend floating point
-HANDLE_CAST_INST(47, PtrToInt, PtrToIntInst)  // Pointer -> Integer
-HANDLE_CAST_INST(48, IntToPtr, IntToPtrInst)  // Integer -> Pointer
-HANDLE_CAST_INST(49, BitCast , BitCastInst )  // Type cast
-HANDLE_CAST_INST(50, AddrSpaceCast, AddrSpaceCastInst)  // addrspace cast
-  LAST_CAST_INST(50)
+HANDLE_CAST_INST(47, PtrToInt, PtrToIntInst)  // Pointer -> Integer (bitcast)
+HANDLE_CAST_INST(48, PtrToAddr, PtrToAddrInst) // Pointer -> Address
+HANDLE_CAST_INST(49, IntToPtr, IntToPtrInst)  // Integer -> Pointer
+HANDLE_CAST_INST(50, BitCast , BitCastInst )  // Type cast
+HANDLE_CAST_INST(51, AddrSpaceCast, AddrSpaceCastInst)  // addrspace cast
+  LAST_CAST_INST(51)
 
- FIRST_FUNCLETPAD_INST(51)
-HANDLE_FUNCLETPAD_INST(51, CleanupPad, CleanupPadInst)
-HANDLE_FUNCLETPAD_INST(52, CatchPad  , CatchPadInst)
-  LAST_FUNCLETPAD_INST(52)
+ FIRST_FUNCLETPAD_INST(52)
+HANDLE_FUNCLETPAD_INST(52, CleanupPad, CleanupPadInst)
+HANDLE_FUNCLETPAD_INST(53, CatchPad  , CatchPadInst)
+  LAST_FUNCLETPAD_INST(53)
 
 // Other operators...
- FIRST_OTHER_INST(53)
-HANDLE_OTHER_INST(53, ICmp   , ICmpInst   )  // Integer comparison instruction
-HANDLE_OTHER_INST(54, FCmp   , FCmpInst   )  // Floating point comparison instr.
-HANDLE_OTHER_INST(55, PHI    , PHINode    )  // PHI node instruction
-HANDLE_OTHER_INST(56, Call   , CallInst   )  // Call a function
-HANDLE_OTHER_INST(57, Select , SelectInst )  // select instruction
-HANDLE_USER_INST (58, UserOp1, Instruction)  // May be used internally in a pass
-HANDLE_USER_INST (59, UserOp2, Instruction)  // Internal to passes only
-HANDLE_OTHER_INST(60, VAArg  , VAArgInst  )  // vaarg instruction
-HANDLE_OTHER_INST(61, ExtractElement, ExtractElementInst)// extract from vector
-HANDLE_OTHER_INST(62, InsertElement, InsertElementInst)  // insert into vector
-HANDLE_OTHER_INST(63, ShuffleVector, ShuffleVectorInst)  // shuffle two vectors.
-HANDLE_OTHER_INST(64, ExtractValue, ExtractValueInst)// extract from aggregate
-HANDLE_OTHER_INST(65, InsertValue, InsertValueInst)  // insert into aggregate
-HANDLE_OTHER_INST(66, LandingPad, LandingPadInst)  // Landing pad instruction.
-HANDLE_OTHER_INST(67, Freeze, FreezeInst) // Freeze instruction.
-  LAST_OTHER_INST(67)
+ FIRST_OTHER_INST(54)
+HANDLE_OTHER_INST(54, ICmp   , ICmpInst   )  // Integer comparison instruction
+HANDLE_OTHER_INST(55, FCmp   , FCmpInst   )  // Floating point comparison instr.
+HANDLE_OTHER_INST(56, PHI    , PHINode    )  // PHI node instruction
+HANDLE_OTHER_INST(57, Call   , CallInst   )  // Call a function
+HANDLE_OTHER_INST(58, Select , SelectInst )  // select instruction
+HANDLE_USER_INST (59, UserOp1, Instruction)  // May be used internally in a pass
+HANDLE_USER_INST (60, UserOp2, Instruction)  // Internal to passes only
+HANDLE_OTHER_INST(61, VAArg  , VAArgInst  )  // vaarg instruction
+HANDLE_OTHER_INST(62, ExtractElement, ExtractElementInst)// extract from vector
+HANDLE_OTHER_INST(63, InsertElement, InsertElementInst)  // insert into vector
+HANDLE_OTHER_INST(64, ShuffleVector, ShuffleVectorInst)  // shuffle two vectors.
+HANDLE_OTHER_INST(65, ExtractValue, ExtractValueInst)// extract from aggregate
+HANDLE_OTHER_INST(66, InsertValue, InsertValueInst)  // insert into aggregate
+HANDLE_OTHER_INST(67, LandingPad, LandingPadInst)  // Landing pad instruction.
+HANDLE_OTHER_INST(68, Freeze, FreezeInst) // Freeze instruction.
+  LAST_OTHER_INST(68)
 
 #undef  FIRST_TERM_INST
 #undef HANDLE_TERM_INST

llvm/include/llvm/IR/Instructions.h

@@ -4949,6 +4949,46 @@ class PtrToIntInst : public CastInst {
   }
 };
 
+/// This class represents a cast from a pointer to an address (non-capturing
+/// ptrtoint).
+class PtrToAddrInst : public CastInst {
+protected:
+  // Note: Instruction needs to be a friend here to call cloneImpl.
+  friend class Instruction;
+
+  /// Clone an identical PtrToAddrInst.
+  PtrToAddrInst *cloneImpl() const;
+
+public:
+  /// Constructor with insert-before-instruction semantics
+  PtrToAddrInst(Value *S,                  ///< The value to be converted
+                Type *Ty,                  ///< The type to convert to
+                const Twine &NameStr = "", ///< A name for the new instruction
+                InsertPosition InsertBefore =
+                    nullptr ///< Where to insert the new instruction
+  );
+
+  /// Gets the pointer operand.
+  Value *getPointerOperand() { return getOperand(0); }
+  /// Gets the pointer operand.
+  const Value *getPointerOperand() const { return getOperand(0); }
+  /// Gets the operand index of the pointer operand.
+  static unsigned getPointerOperandIndex() { return 0U; }
+
+  /// Returns the address space of the pointer operand.
+  unsigned getPointerAddressSpace() const {
+    return getPointerOperand()->getType()->getPointerAddressSpace();
+  }
+
+  // Methods for support type inquiry through isa, cast, and dyn_cast:
+  static bool classof(const Instruction *I) {
+    return I->getOpcode() == PtrToAddr;
+  }
+  static bool classof(const Value *V) {
+    return isa<Instruction>(V) && classof(cast<Instruction>(V));
+  }
+};
+
 //===----------------------------------------------------------------------===//
 //                             BitCastInst Class
 //===----------------------------------------------------------------------===//

llvm/include/llvm/IR/Operator.h

@@ -595,6 +595,37 @@ struct OperandTraits<PtrToIntOperator>
 
 DEFINE_TRANSPARENT_OPERAND_ACCESSORS(PtrToIntOperator, Value)
 
+class PtrToAddrOperator
+    : public ConcreteOperator<Operator, Instruction::PtrToAddr> {
+  friend class PtrToAddr;
+  friend class ConstantExpr;
+
+public:
+  /// Transparently provide more efficient getOperand methods.
+  DECLARE_TRANSPARENT_OPERAND_ACCESSORS(Value);
+
+  Value *getPointerOperand() { return getOperand(0); }
+  const Value *getPointerOperand() const { return getOperand(0); }
+
+  static unsigned getPointerOperandIndex() {
+    return 0U; // get index for modifying correct operand
+  }
+
+  /// Method to return the pointer operand as a PointerType.
+  Type *getPointerOperandType() const { return getPointerOperand()->getType(); }
+
+  /// Method to return the address space of the pointer operand.
+  unsigned getPointerAddressSpace() const {
+    return cast<PointerType>(getPointerOperandType())->getAddressSpace();
+  }
+};
+
+template <>
+struct OperandTraits<PtrToAddrOperator>
+    : public FixedNumOperandTraits<PtrToAddrOperator, 1> {};
+
+DEFINE_TRANSPARENT_OPERAND_ACCESSORS(PtrToAddrOperator, Value)
+
 class BitCastOperator
     : public ConcreteOperator<Operator, Instruction::BitCast> {
   friend class BitCastInst;

llvm/include/llvm/SandboxIR/Instruction.h

@@ -2278,6 +2278,8 @@ class CastInst : public UnaryInstruction {
       return Opcode::FPToSI;
     case llvm::Instruction::FPExt:
       return Opcode::FPExt;
+    case llvm::Instruction::PtrToAddr:
+      return Opcode::PtrToAddr;
     case llvm::Instruction::PtrToInt:
       return Opcode::PtrToInt;
     case llvm::Instruction::IntToPtr:
@@ -2364,6 +2366,8 @@ class FPToUIInst final : public CastInstImpl<Instruction::Opcode::FPToUI> {};
 class FPToSIInst final : public CastInstImpl<Instruction::Opcode::FPToSI> {};
 class IntToPtrInst final : public CastInstImpl<Instruction::Opcode::IntToPtr> {
 };
+class PtrToAddrInst final
+    : public CastInstImpl<Instruction::Opcode::PtrToAddr> {};
 class PtrToIntInst final : public CastInstImpl<Instruction::Opcode::PtrToInt> {
 };
 class BitCastInst final : public CastInstImpl<Instruction::Opcode::BitCast> {};

llvm/include/llvm/SandboxIR/Values.def

@@ -118,6 +118,7 @@ DEF_INSTR(Cast,   OPCODES(\
                           OP(FPToUI)        \
                           OP(FPToSI)        \
                           OP(FPExt)         \
+                          OP(PtrToAddr)     \
                           OP(PtrToInt)      \
                           OP(IntToPtr)      \
                           OP(SIToFP)        \

llvm/lib/Analysis/ConstantFolding.cpp

@@ -1482,6 +1482,20 @@ Constant *llvm::ConstantFoldCastOperand(unsigned Opcode, Constant *C,
   switch (Opcode) {
   default:
     llvm_unreachable("Missing case");
+  case Instruction::PtrToAddr:
+    if (auto *GEP = dyn_cast<GEPOperator>(C)) {
+      // For now just handle the basic case of GEPs on NULL for ptrtoaddr.
+      // (ptrtoaddr (gep null, x)) -> x
+      // (ptrtoaddr (gep (gep null, x), y) -> x + y, etc.
+      unsigned BitWidth = DL.getIndexTypeSizeInBits(GEP->getType());
+      APInt BaseOffset(BitWidth, 0);
+      auto *Base = cast<Constant>(GEP->stripAndAccumulateConstantOffsets(
+          DL, BaseOffset, /*AllowNonInbounds=*/true));
+      if (Base->isNullValue()) {
+        return ConstantInt::get(C->getContext(), BaseOffset);
+      }
+    }
+    break;
   case Instruction::PtrToInt:
     if (auto *CE = dyn_cast<ConstantExpr>(C)) {
       Constant *FoldedValue = nullptr;

llvm/lib/AsmParser/LLLexer.cpp

@@ -928,6 +928,7 @@ lltok::Kind LLLexer::LexIdentifier() {
   INSTKEYWORD(fptoui,      FPToUI);
   INSTKEYWORD(fptosi,      FPToSI);
   INSTKEYWORD(inttoptr,    IntToPtr);
+  INSTKEYWORD(ptrtoaddr,   PtrToAddr);
   INSTKEYWORD(ptrtoint,    PtrToInt);
   INSTKEYWORD(bitcast,     BitCast);
   INSTKEYWORD(addrspacecast, AddrSpaceCast);

llvm/lib/AsmParser/LLParser.cpp

@@ -4273,6 +4273,7 @@ bool LLParser::parseValID(ValID &ID, PerFunctionState *PFS, Type *ExpectedTy) {
   case lltok::kw_bitcast:
   case lltok::kw_addrspacecast:
   case lltok::kw_inttoptr:
+  case lltok::kw_ptrtoaddr:
   case lltok::kw_ptrtoint: {
     unsigned Opc = Lex.getUIntVal();
     Type *DestTy = nullptr;
@@ -7310,6 +7311,7 @@ int LLParser::parseInstruction(Instruction *&Inst, BasicBlock *BB,
   case lltok::kw_fptoui:
   case lltok::kw_fptosi:
   case lltok::kw_inttoptr:
+  case lltok::kw_ptrtoaddr:
   case lltok::kw_ptrtoint:
     return parseCast(Inst, PFS, KeywordVal);
   case lltok::kw_fptrunc: