[IR] Add a target extension type to LLVM.

Target-extension types represent types that need to be preserved through
optimization, but otherwise are not introspectable by target-independent
optimizations. This patch doesn't add any uses of these types by an existing
backend, it only provides basic infrastructure such that these types would work
correctly.

Reviewed By: nikic, barannikov88

Differential Revision: https://reviews.llvm.org/D135202

Author: jcranmer-intel

Diff for: llvm/docs/BitCodeFormat.rst

@@ -1338,6 +1338,21 @@ TYPE_CODE_X86_AMX Record
 
 The ``X86_AMX`` record (code 24) adds an ``x86_amx`` type to the type table.
 
+TYPE_CODE_TARGET_TYPE Record
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+``[TARGET_TYPE, num_tys, ...ty_params..., ...int_params... ]``
+
+The ``TARGET_TYPE`` record (code 26) adds a target extension type to the type
+table, with a name defined by a previously encountered ``STRUCT_NAME`` record.
+The operand fields are
+
+* *num_tys*: The number of parameters that are types (as opposed to integers)
+
+* *ty_params*: Type indices that represent type parameters
+
+* *int_params*: Numbers that correspond to the integer parameters.
+
 .. _CONSTANTS_BLOCK:
 
 CONSTANTS_BLOCK Contents

Diff for: llvm/docs/LangRef.rst

@@ -3631,6 +3631,53 @@ Prior to LLVM 15, pointer types also specified a pointee type, such as
 pointers" are still supported under non-default options. See the
 `opaque pointers document <OpaquePointers.html>`__ for more information.
 
+.. _t_target_type:
+
+Target Extension Type
+"""""""""""""""""""""
+
+:Overview:
+
+Target extension types represent types that must be preserved through
+optimization, but are otherwise generally opaque to the compiler. They may be
+used as function parameters or arguments, and in :ref:`phi <i_phi>` or
+:ref:`select <i_select>` instructions. Some types may be also used in
+:ref:`alloca <i_alloca>` instructions or as global values, and correspondingly
+it is legal to use :ref:`load <i_load>` and :ref:`store <i_store>` instructions
+on them. Full semantics for these types are defined by the target.
+
+The only constants that target extension types may have are ``zeroinitializer``,
+``undef``, and ``poison``. Other possible values for target extension types may
+arise from target-specific intrinsics and functions.
+
+These types cannot be converted to other types. As such, it is not legal to use
+them in :ref:`bitcast <i_bitcast>` instructions (as a source or target type),
+nor is it legal to use them in :ref:`ptrtoint <i_ptrtoint>` or
+:ref:`inttoptr <i_inttoptr>` instructions. Similarly, they are not legal to use
+in an :ref:`icmp <i_icmp>` instruction.
+
+Target extension types have a name and optional type or integer parameters. The
+meanings of name and parameters are defined by the target. When being defined in
+LLVM IR, all of the type parameters must precede all of the integer parameters.
+
+Specific target extension types are registered with LLVM as having specific
+properties. These properties can be used to restrict the type from appearing in
+certain contexts, such as being the type of a global variable or having a
+``zeroinitializer`` constant be valid. A complete list of type properties may be
+found in the documentation for ``llvm::TargetExtType::Property`` (`doxygen
+<https://llvm.org/doxygen/classllvm_1_1TargetExtType.html>`_).
+
+:Syntax:
+
+.. code-block:: llvm
+
+      target("label")
+      target("label", void)
+      target("label", void, i32)
+      target("label", 0, 1, 2)
+      target("label", void, i32, 0, 1, 2)
+
+
 .. _t_vector:
 
 Vector Type

Diff for: llvm/docs/ReleaseNotes.rst

@@ -104,6 +104,10 @@ Changes to the LLVM IR
 
   * ``fneg``
 
+* Target extension types have been added, which allow targets to have
+  types that need to be preserved through the optimizer, but otherwise are not
+  introspectable by target-independent optimizations.
+
 Changes to building LLVM
 ------------------------
 

Diff for: llvm/include/llvm-c/Core.h

@@ -165,7 +165,8 @@ typedef enum {
   LLVMTokenTypeKind,     /**< Tokens */
   LLVMScalableVectorTypeKind, /**< Scalable SIMD vector type */
   LLVMBFloatTypeKind,    /**< 16 bit brain floating point type */
-  LLVMX86_AMXTypeKind    /**< X86 AMX */
+  LLVMX86_AMXTypeKind,   /**< X86 AMX */
+  LLVMTargetExtTypeKind, /**< Target extension type */
 } LLVMTypeKind;
 
 typedef enum {
@@ -284,7 +285,8 @@ typedef enum {
   LLVMInlineAsmValueKind,
 
   LLVMInstructionValueKind,
-  LLVMPoisonValueValueKind
+  LLVMPoisonValueValueKind,
+  LLVMConstantTargetNoneValueKind,
 } LLVMValueKind;
 
 typedef enum {
@@ -1571,6 +1573,15 @@ LLVMTypeRef LLVMLabelType(void);
 LLVMTypeRef LLVMX86MMXType(void);
 LLVMTypeRef LLVMX86AMXType(void);
 
+/**
+ * Create a target extension type in LLVM context.
+ */
+LLVMTypeRef LLVMTargetExtTypeInContext(LLVMContextRef C, const char *Name,
+                                       LLVMTypeRef *TypeParams,
+                                       unsigned TypeParamCount,
+                                       unsigned *IntParams,
+                                       unsigned IntParamCount);
+
 /**
  * @}
  */

Diff for: llvm/include/llvm/AsmParser/LLParser.h

@@ -433,6 +433,7 @@ namespace llvm {
 
     bool parseArrayVectorType(Type *&Result, bool IsVector);
     bool parseFunctionType(Type *&Result);
+    bool parseTargetExtType(Type *&Result);
 
     // Function Semantic Analysis.
     class PerFunctionState {

Diff for: llvm/include/llvm/Bitcode/LLVMBitCodes.h

@@ -175,6 +175,8 @@ enum TypeCodes {
   TYPE_CODE_X86_AMX = 24, // X86 AMX
 
   TYPE_CODE_OPAQUE_POINTER = 25, // OPAQUE_POINTER: [addrspace]
+
+  TYPE_CODE_TARGET_TYPE = 26, // TARGET_TYPE
 };
 
 enum OperandBundleTagCode {

Diff for: llvm/include/llvm/IR/Constants.h

@@ -843,6 +843,33 @@ class ConstantTokenNone final : public ConstantData {
   }
 };
 
+/// A constant target extension type default initializer
+class ConstantTargetNone final : public ConstantData {
+  friend class Constant;
+
+  explicit ConstantTargetNone(TargetExtType *T)
+      : ConstantData(T, Value::ConstantTargetNoneVal) {}
+
+  void destroyConstantImpl();
+
+public:
+  ConstantTargetNone(const ConstantTargetNone &) = delete;
+
+  /// Static factory methods - Return objects of the specified value.
+  static ConstantTargetNone *get(TargetExtType *T);
+
+  /// Specialize the getType() method to always return an TargetExtType,
+  /// which reduces the amount of casting needed in parts of the compiler.
+  inline TargetExtType *getType() const {
+    return cast<TargetExtType>(Value::getType());
+  }
+
+  /// Methods for support type inquiry through isa, cast, and dyn_cast.
+  static bool classof(const Value *V) {
+    return V->getValueID() == ConstantTargetNoneVal;
+  }
+};
+
 /// The address of a basic block.
 ///
 class BlockAddress final : public Constant {

Diff for: llvm/include/llvm/IR/DataLayout.h

@@ -713,6 +713,10 @@ inline TypeSize DataLayout::getTypeSizeInBits(Type *Ty) const {
                        getTypeSizeInBits(VTy->getElementType()).getFixedSize();
     return TypeSize(MinBits, EltCnt.isScalable());
   }
+  case Type::TargetExtTyID: {
+    Type *LayoutTy = cast<TargetExtType>(Ty)->getLayoutType();
+    return getTypeSizeInBits(LayoutTy);
+  }
   default:
     llvm_unreachable("DataLayout::getTypeSizeInBits(): Unsupported type");
   }

Diff for: llvm/include/llvm/IR/DerivedTypes.h

@@ -730,6 +730,82 @@ unsigned Type::getPointerAddressSpace() const {
   return cast<PointerType>(getScalarType())->getAddressSpace();
 }
 
+/// Class to represent target extensions types, which are generally
+/// unintrospectable from target-independent optimizations.
+///
+/// Target extension types have a string name, and optionally have type and/or
+/// integer parameters. The exact meaning of any parameters is dependent on the
+/// target.
+class TargetExtType : public Type {
+  TargetExtType(LLVMContext &C, StringRef Name, ArrayRef<Type *> Types,
+                ArrayRef<unsigned> Ints);
+
+  std::string Name;
+  unsigned *IntParams;
+
+public:
+  TargetExtType(const TargetExtType &) = delete;
+  TargetExtType &operator=(const TargetExtType &) = delete;
+
+  /// Return a target extension type having the specified name and optional
+  /// type and integer parameters.
+  static TargetExtType *get(LLVMContext &Context, StringRef Name,
+                            ArrayRef<Type *> Types = std::nullopt,
+                            ArrayRef<unsigned> Ints = std::nullopt);
+
+  /// Return the name for this target extension type. Two distinct target
+  /// extension types may have the same name if their type or integer parameters
+  /// differ.
+  StringRef getName() const { return Name; }
+
+  /// Return the type parameters for this particular target extension type. If
+  /// there are no parameters, an empty array is returned.
+  ArrayRef<Type *> type_params() const {
+    return makeArrayRef(type_param_begin(), type_param_end());
+  }
+
+  using type_param_iterator = Type::subtype_iterator;
+  type_param_iterator type_param_begin() const { return ContainedTys; }
+  type_param_iterator type_param_end() const {
+    return &ContainedTys[NumContainedTys];
+  }
+
+  Type *getTypeParameter(unsigned i) const { return getContainedType(i); }
+  unsigned getNumTypeParameters() const { return getNumContainedTypes(); }
+
+  /// Return the integer parameters for this particular target extension type.
+  /// If there are no parameters, an empty array is returned.
+  ArrayRef<unsigned> int_params() const {
+    return makeArrayRef(IntParams, getNumIntParameters());
+  }
+
+  unsigned getIntParameter(unsigned i) const { return IntParams[i]; }
+  unsigned getNumIntParameters() const { return getSubclassData(); }
+
+  enum Property {
+    /// zeroinitializer is valid for this target extension type.
+    HasZeroInit = 1U << 0,
+    /// This type may be used as the value type of a global variable.
+    CanBeGlobal = 1U << 1,
+  };
+
+  /// Returns true if the target extension type contains the given property.
+  bool hasProperty(Property Prop) const;
+
+  /// Returns an underlying layout type for the target extension type. This
+  /// type can be used to query size and alignment information, if it is
+  /// appropriate (although note that the layout type may also be void). It is
+  /// not legal to bitcast between this type and the layout type, however.
+  Type *getLayoutType() const;
+
+  /// Methods for support type inquiry through isa, cast, and dyn_cast.
+  static bool classof(const Type *T) { return T->getTypeID() == TargetExtTyID; }
+};
+
+StringRef Type::getTargetExtName() const {
+  return cast<TargetExtType>(this)->getName();
+}
+
 } // end namespace llvm
 
 #endif // LLVM_IR_DERIVEDTYPES_H

Diff for: llvm/include/llvm/IR/Type.h

@@ -76,6 +76,7 @@ class Type {
     FixedVectorTyID,    ///< Fixed width SIMD vector type
     ScalableVectorTyID, ///< Scalable SIMD vector type
     TypedPointerTyID,   ///< Typed pointer used by some GPU targets
+    TargetExtTyID,      ///< Target extension type
   };
 
 private:
@@ -194,6 +195,9 @@ class Type {
   /// Return true if this is X86 AMX.
   bool isX86_AMXTy() const { return getTypeID() == X86_AMXTyID; }
 
+  /// Return true if this is a target extension type.
+  bool isTargetExtTy() const { return getTypeID() == TargetExtTyID; }
+
   /// Return true if this is a FP type or a vector of FP.
   bool isFPOrFPVectorTy() const { return getScalarType()->isFloatingPointTy(); }
 
@@ -267,7 +271,7 @@ class Type {
   /// includes all first-class types except struct and array types.
   bool isSingleValueType() const {
     return isFloatingPointTy() || isX86_MMXTy() || isIntegerTy() ||
-           isPointerTy() || isVectorTy() || isX86_AMXTy();
+           isPointerTy() || isVectorTy() || isX86_AMXTy() || isTargetExtTy();
   }
 
   /// Return true if the type is an aggregate type. This means it is valid as
@@ -288,7 +292,8 @@ class Type {
       return true;
     // If it is not something that can have a size (e.g. a function or label),
     // it doesn't have a size.
-    if (getTypeID() != StructTyID && getTypeID() != ArrayTyID && !isVectorTy())
+    if (getTypeID() != StructTyID && getTypeID() != ArrayTyID &&
+        !isVectorTy() && getTypeID() != TargetExtTyID)
       return false;
     // Otherwise we have to try harder to decide.
     return isSizedDerivedType(Visited);
@@ -386,6 +391,8 @@ class Type {
     return ContainedTys[0];
   }
 
+  inline StringRef getTargetExtName() const;
+
   /// This method is deprecated without replacement. Pointer element types are
   /// not available with opaque pointers.
   [[deprecated("Deprecated without replacement, see "

Diff for: llvm/include/llvm/IR/Value.def

@@ -95,6 +95,7 @@ HANDLE_CONSTANT(ConstantDataArray)
 HANDLE_CONSTANT(ConstantDataVector)
 HANDLE_CONSTANT(ConstantInt)
 HANDLE_CONSTANT(ConstantFP)
+HANDLE_CONSTANT(ConstantTargetNone)
 HANDLE_CONSTANT(ConstantPointerNull)
 HANDLE_CONSTANT(ConstantTokenNone)