10.9 Automatic memory management
C# employs automatic memory management, which frees developers from manually
allocating and freeing
the memory occupied by objects. Automatic memory management policies are
implemented by a garbage
collector. The memory management life cycle of an object is as follows:
1. When the object is created, memory is allocated for it, the constructor
is run, and the object is
considered live.
2. If the object, or any part of it, cannot be accessed by any possible
continuation of execution, other than
the running of destructors, the object is considered no longer in use, and
it becomes eligible for
destruction. [Note: Implementations may choose to analyze code to determine
which references to an
object may be used in the future. For instance, if a local variable that is
in scope is the only existing
reference to an object, but that local variable is never referred to in any
possible continuation of
execution from the current execution point in the procedure, an
implementation may (but is not required
to) treat the object as no longer in use. end note]
3. Once the object is eligible for destruction, at some unspecified later
time the destructor (§17.12) (if any)
for the object is run. Unless overridden by explicit calls, the destructor
for the object is run once only.
4. Once the destructor for an object is run, if that object, or any part of
it, cannot be accessed by any
possible continuation of execution, including the running of destructors,
the object is considered
inaccessible and the object becomes eligible for collection.
5. Finally, at some time after the object becomes eligible for collection,
the garbage collector frees the
memory associated with that object.
The garbage collector maintains information about object usage, and uses
this information to make memory
management decisions, such as where in memory to locate a newly created
object, when to relocate an
object, and when an object is no longer in use or inaccessible.
Like other languages that assume the existence of a garbage collector, C#
is designed so that the garbage
collector may implement a wide range of memory management policies. For
instance, C# does not require
that destructors be run or that objects be collected as soon as they are
eligible, or that destructors be run in
any particular order, or on any particular thread.
The behavior of the garbage collector can be controlled, to some degree,
via static methods on the class
System.GC. [Note: While not required by this specification, a Collect
method may be provided on the GC
class which requests that a collection be performed. The following examples
presume the existence of such
a method. end note]
[Example: Since the garbage collector is allowed wide latitude in deciding
when to collect objects and run
destructors, a conforming implementation may produce output that differs
from that shown by the following
code. The program
using System;
class A
{
~A() {
Console.WriteLine("Destruct instance of A");
}
}
C# LANGUAGE SPECIFICATION
86
class B
{
object Ref;
public B(object o) {
Ref = o;
}
~B() {
Console.WriteLine("Destruct instance of B");
}
}
class Test
{
static void Main() {
B b = new B(new A());
b = null;
GC.Collect();
GC.WaitForPendingFinalizers();
}
}
creates an instance of class A and an instance of class B. These objects
become eligible for garbage
collection when the variable b is assigned the value null, since after this
time it is impossible for any userwritten
code to access them. The output could be either
Destruct instance of A
Destruct instance of B
or
Destruct instance of B
Destruct instance of A
because the language imposes no constraints on the order in which objects
are garbage collected.
In subtle cases, the distinction between .eligible for destruction. and
.eligible for collection. can be
important. For example,
using System;
class A
{
~A() {
Console.WriteLine("Destruct instance of A");
}
public void F() {
Console.WriteLine("A.F");
Test.RefA = this;
}
}
class B
{
public A Ref;
~B() {
Console.WriteLine("Destruct instance of B");
Ref.F();
}
}
class Test
{
public static A RefA;
public static B RefB;
static void Main() {
RefB = new B();
RefA = new A();
RefB.Ref = RefA;
RefB = null;
RefA = null;
Chapter 10 Basic concepts
87
// A and B now eligible for destruction
GC.Collect();
GC.WaitForPendingFinalizers();
// B now eligible for collection, but A is not
if (RefA != null)
Console.WriteLine("RefA is not null");
}
}
In the above program, if the garbage collector chooses to run the
destructor of A before the destructor of B,
then the output of this program might be:
Destruct instance of A
Destruct instance of B
A.F
RefA is not null
Note that although the instance of A was not in use and A’s destructor was
run, it is still possible for methods
of A (in this case, F) to be called from another destructor. Also, note
that running of a destructor may cause
an object to become usable from the mainline program again. In this case,
the running of B’s destructor
caused an instance of A that was previously not in use to become accessible
from the live reference RefA.
After the call to WaitForPendingFinalizers, the instance of B is eligible
for collection, but the instance
of A is not, because of the reference RefA.
To avoid confusion and unexpected behavior, it is generally a good idea for
destructors to only perform
cleanup on data stored in their object’s own fields, and not to perform
any actions on referenced objects or
static fields. end example]