Holding locks while performing time-consuming or blocking operations can severely degrade system performance and can result in starvation. Furthermore, deadlock can result if interdependent threads block indefinitely. Blocking operations include network, file, and console I/O (for example, Console.readLine()
) and object serialization. Deferring a thread indefinitely also constitutes a blocking operation. Consequently, programs must not perform blocking operations while holding a lock.
If the JVM interacts with a file system that operates over an unreliable network, file I/O might incur a large performance penalty. In such cases, avoid file I/O over the network when holding a lock. File operations (such as logging) that could block waiting for the output stream lock or for I/O to complete could be performed in a dedicated thread to speed up task processing. Logging requests can be added to a queue, assuming that the queue's put()
operation incurs little overhead as compared to file I/O [[Goetz 2006]].
Noncompliant Code Example (Deferring a Thread)
This noncompliant code example defines a utility method that accepts a time
argument.
public synchronized void doSomething(long time) throws InterruptedException { // ... Thread.sleep(time); }
Because the method is synchronized, when the thread is suspended, other threads are unable to use the synchronized methods of the class. The current object's monitor continues to be held because the Thread.sleep()
method lacks synchronization semantics, as detailed in rule THI00-J. Do not assume that the sleep(), yield() or getState() methods provide synchronization semantics.
Compliant Solution (Intrinsic Lock)
This compliant solution defines the doSomething()
method with a timeout
parameter rather than the time
value. Using Object.wait()
instead of Thread.sleep()
allows setting a time out period during which a notification may awaken the thread.
public synchronized void doSomething(long timeout) throws InterruptedException { while (<condition does not hold>) { wait(timeout); // Immediately releases current monitor } }
The current object's monitor is immediately released upon entering the wait state. After the time out period has elapsed, the thread resumes execution after reacquiring the current object's monitor.
According to the Java API class Object
documentation [[API 2006]]
Note that the
wait
method, as it places the current thread into the wait set for this object, unlocks only this object; any other objects on which the current thread may be synchronized remain locked while the thread waits. This method should only be called by a thread that is the owner of this object's monitor.
Programs must ensure that threads that hold locks on other objects release those locks appropriately, before entering the wait state. Additional guidance on waiting and notification is available in rules THI03-J. Always invoke wait() and await() methods inside a loop and THI02-J. Notify all waiting threads rather than a single thread.
Noncompliant Code Example (Network I/O)
This noncompliant code example shows a sendPage()
method that sends a Page
object from a server to a client. The method is synchronized to protect the pageBuff
array when multiple threads request concurrent access.
// Class Page is defined separately. It stores and returns the Page name via getName() Page[] pageBuff = new Page[MAX_PAGE_SIZE]; public synchronized boolean sendPage(Socket socket, String pageName) throws IOException { // Get the output stream to write the Page to ObjectOutputStream out = new ObjectOutputStream(socket.getOutputStream()); // Find the Page requested by the client (this operation requires synchronization) Page targetPage = null; for (Page p : pageBuff) { if (p.getName().compareTo(pageName) == 0) { targetPage = p; } } // Requested Page does not exist if (targetPage == null) { return false; } // Send the Page to the client (does not require any synchronization) out.writeObject(targetPage); out.flush(); out.close(); return true; }
Calling writeObject()
within the synchronized sendPage()
method can result in delays and deadlock-like conditions in high latency networks or when network connections are inherently lossy.
Compliant Solution
This compliant solution separates the process into a sequence of steps:
- Perform actions on data structures requiring synchronization.
- Create copies of the objects to be sent.
- Perform network calls in a separate unsynchronized method.
In this compliant solution, the synchronized getPage()
method is called from an unsynchronized sendPage()
method to retrieve the requested Page
in the pageBuff
array. After the Page
is retrieved, sendPage()
calls the unsynchronized deliverPage()
method to deliver the Page
to the client.
public boolean sendPage(Socket socket, String pageName) { // No synchronization Page targetPage = getPage(pageName); if (targetPage == null) return false; return deliverPage(socket, targetPage); } private synchronized Page getPage(String pageName) { // Requires synchronization Page targetPage = null; for (Page p : pageBuff) { if (p.getName().equals(pageName)) { targetPage = p; } } return targetPage; } // Return false if an error occurs, true if successful public boolean deliverPage(Socket socket, Page page) { ObjectOutputStream out = null; boolean result = true; try { // Get the output stream to write the Page to out = new ObjectOutputStream(socket.getOutputStream()); // Send the Page to the client out.writeObject(page); } catch (IOException io) { result = false; } finally { if (out != null) { try { out.flush(); out.close(); } catch (IOException e) { result = false; } } } return result; }
Exceptions
LCK09-EX0: Classes that provide an appropriate termination mechanism to callers are permitted to violate this rule. See rule THI04-J. Ensure that threads performing blocking operations can be terminated.
LCK09-EX1: Method that require multiple locks may hold several locks while waiting for the remaining locks to become available. This constitutes a valid exception, although the programmer must follow other applicable rules to avoid deadlock. See rule LCK07-J. Avoid deadlock by requesting and releasing locks in the same order for more information.
Risk Assessment
Blocking or lengthy operations performed within synchronized regions could result in a deadlocked or unresponsive system.
Rule |
Severity |
Likelihood |
Remediation Cost |
Priority |
Level |
---|---|---|---|---|---|
LCK09-J |
low |
probable |
high |
P2 |
L3 |
Related Guidelines
Bibliography
<ac:structured-macro ac:name="unmigrated-wiki-markup" ac:schema-version="1" ac:macro-id="98b9e2cd-3240-4014-8ebf-d1314b2a5e41"><ac:plain-text-body><![CDATA[ |
[[API 2006 |
AA. Bibliography#API 06]] |
Class |
]]></ac:plain-text-body></ac:structured-macro> |
|
<ac:structured-macro ac:name="unmigrated-wiki-markup" ac:schema-version="1" ac:macro-id="7cd06f30-1e26-4043-a5d2-6be77404bcea"><ac:plain-text-body><![CDATA[ |
[[Grosso 2001 |
AA. Bibliography#Grosso 01]] |
[Chapter 10: Serialization |
http://oreilly.com/catalog/javarmi/chapter/ch10.html] |
]]></ac:plain-text-body></ac:structured-macro> |
<ac:structured-macro ac:name="unmigrated-wiki-markup" ac:schema-version="1" ac:macro-id="923e47b9-c9f4-423f-b97a-13af89e84395"><ac:plain-text-body><![CDATA[ |
[[JLS 2005 |
AA. Bibliography#JLS 05]] |
[Chapter 17, Threads and Locks |
http://java.sun.com/docs/books/jls/third_edition/html/memory.html] |
]]></ac:plain-text-body></ac:structured-macro> |
<ac:structured-macro ac:name="unmigrated-wiki-markup" ac:schema-version="1" ac:macro-id="7d5c715b-6e63-4e08-a2cd-5ae58f95a7a3"><ac:plain-text-body><![CDATA[ |
[[Rotem 2008 |
AA. Bibliography#Rotem 08]] |
[Falacies of Distributed Computing Explained |
http://www.rgoarchitects.com/Files/fallacies.pdf] |
]]></ac:plain-text-body></ac:structured-macro> |
08. Locking (LCK) LCK10-J. Do not use incorrect forms of the double-checked locking idiom