C# Common Type System

The C# typing system contains the following categories:

Value types

Reference types

Pointer types

Variables that are value types store data, and those that are reference types store references to the actual data. Reference types are also referred to as objects. Pointer types can be used only in unsafe mode.

It is possible to convert a value type to a reference type, and back again to a value type, by using boxing and unboxing. With the exception of a boxed value type, you cannot convert a reference type to a value type.

Types can derive from other types, called base types. The derived type inherits (with some restrictions) the methods, properties, and other members of the base type. The base type can in turn derive from some other type, in which case the derived type inherits the members of both base types in its inheritance hierarchy. All types, including built-in numeric types such as System.Int32 (C# keyword: int), derive ultimately from a single base type, which is System.Object (C# keyword: object). This unified type hierarchy is called the Common Type System (CTS).

Each type in the CTS is defined as either a value type or a reference type. This includes all custom types in the .NET Framework class library and also your own user-defined types. Types that you define by using the struct keyword are value types; all the built-in numeric types are structs. Types that you define by using the class keyword are reference types. Reference types and value types have different compile-time rules, and different run-time behavior.

The following illustration shows the relationship between value types and reference types in the CTS.

Value Types

Value types derive from System.ValueType, which derives from System.Object. Types that derive from System.ValueType have special behavior in the CLR. Value type variables directly contain their values, which means that the memory is allocated inline in whatever context the variable is declared. There is no separate heap allocation or garbage collection overhead for value-type variables.

There are two categories of value types: struct and enum.

Value types are sealed, which means, for example, that you cannot derive a type from System..::.Int32, and you cannot define a struct to inherit from any user-defined class or struct because a struct can only inherit from System.ValueType. However, a struct can implement one or more interfaces. You can cast a struct type to an interface type; this causes a boxing operation to wrap the struct inside a reference type object on the managed heap. Boxing operations occur when you pass a value type to a method that takes a System.Object as an input parameter.

Reference Types

A type that is defined as a class, delegate, array, or interface is a reference type. At run time, when you declare a variable of a reference type, the variable contains the value null until you explicitly create an instance of the object by using the new operator, or assign it an object that has been created elsewhere by using new.

When the object is created, the memory is allocated on the managed heap, and the variable holds only a reference to the location of the object. Types on the managed heap require overhead both when they are allocated and when they are reclaimed by the automatic memory management functionality of the CLR, which is known as garbage collection. However, garbage collection is also highly optimized, and in most scenarios it does not create a performance issue.