Reads of Reading a shared primitive variables variable in some one thread may not observe the latest writes to the variables from other threads. It is important to ensure that the accesses see the values yield the value of the most recent writes. If this is not done, multiple threads may observe stale values of the shared variables and fail to act accordingly. Visibility of latest values can be ensured by declaring variables volatile
or correctly synchronizing the code.write to the variable from another thread. Consequently, the thread may observe a stale value of the shared variable. To ensure the visibility of the most recent update, either the variable must be declared volatile or the reads and writes must be synchronized.
Declaring a shared variable volatile guarantees visibility in a thread-safe manner only when both of the following conditions are metThe use of volatile
is recommended under a very restrictive set of conditions:
- A write to a variable does not depend on is independent from its current value.
- The A write is not involved with writes of other variables.
- Only a single thread ever updates the value.
- Locking is not required for any other reason.
- to a variable is independent from the result of any nonatomic compound operations involving reads and writes of other variables (see VNA02-J. Ensure that compound operations on shared variables are atomic for more information).
The first condition can be relaxed when you can be sure that only one thread will ever update the value of the variable [Goetz 2006]. However, code that relies on a single-thread confinement is error prone and difficult to maintain. This design approach is permitted under this rule but is discouraged.
Synchronizing the code makes it easier to reason about its behavior and is frequently more secure than simply using the volatile
keyword. However, synchronization has somewhat higher performance overhead and can result in thread contention and deadlocks when used excessively.
Declaring a variable Furthermore, declaring a variable as volatile or correctly synchronizing the code guarantees that 64-bit primitive variables of type long
and double
variables are accessed atomically (see CON25. For more information on sharing those variables among multiple threads, see VNA05-J. Ensure atomicity when reading and writing 64-bit values for information on sharing long
and double
variables among multiple threads).
Noncompliant Code Example (Non-volatile Flag)
This noncompliant code example uses a shutdown()
method to set a non-volatile the nonvolatile done
flag that is checked in the run()
method. If one thread invokes the shutdown()
method to set the flag, it is possible that another thread might not observe this change. Consequently, the second thread may still observe that done
is false
and incorrectly invoke the sleep()
method.:
Code Block | ||
---|---|---|
| ||
final class ControlledStop implements Runnable { private boolean done = false; @Override public void run() { while (!done) { try { // ... Thread.currentThread().sleep(1000); // Do something } catch(InterruptedException ie) { Thread.currentThread().interrupt(); // Reset handleinterrupted exceptionstatus } } } protectedpublic void shutdown() { done = true; } } |
If one thread invokes the shutdown()
method to set the flag, a second thread might not observe that change. Consequently, the second thread might observe that done
is still false and incorrectly invoke the sleep()
method. Compilers and just-in-time compilers (JITs) are allowed to optimize the code when they determine that the value of done
is never modified by the same thread, resulting in an infinite loop.
Compliant Solution (
...
Volatile
)
In this compliant solution, the done
flag is declared volatile so to ensure that updates writes are visible to other threads.:
Code Block | ||
---|---|---|
| ||
final class ControlledStop implements Runnable { private volatile boolean done = false; @Override public void run() { while (!done) { try { // ... Thread.currentThread().sleep(1000); // Do something } catch(InterruptedException ie) { Thread.currentThread().interrupt(); // Reset handleinterrupted exceptionstatus } } } protectedpublic void shutdown() { done = true; } } |
Compliant Solution (
...
AtomicBoolean
)
This In this compliant solution uses the intrinsic lock of the Class
object to ensure that updates , the done
flag is declared to be of type java.util.concurrent.atomic.AtomicBoolean
. Atomic types also guarantee that writes are visible to other threads.
Code Block | ||
---|---|---|
| ||
final class ControlledStop implements Runnable { private booleanfinal AtomicBoolean done = new AtomicBoolean(false); @Override public void run() { while (!isDonedone.get()) { try { // ... Thread.currentThread().sleep(1000); // Do something } catch(InterruptedException ie) { Thread.currentThread().interrupt(); // Reset handleinterrupted exceptionstatus } } } protected synchronized boolean isDone() { return done; } protected synchronized public void shutdown() { done = true.set(true); } } |
While this is an acceptable compliant solution, it has the following shortcomings as compared to declaring done
as volatile
:
- Performance: The intrinsic locks cause threads to block temporarily;
volatile
incurs no blocking - Deadlock: Excessive synchronization can make the program deadlock prone.
...
Compliant Solution (
...
synchronized
)
This compliant solution uses an AtomicBoolean
flag the intrinsic lock of the Class
object to ensure that updates are visible to other threads.:
Code Block | ||
---|---|---|
| ||
final class ControlledStop implements Runnable { private AtomicBooleanboolean done = new AtomicBoolean(false); @Override public void run() { while (!done.getisDone()) { try { // ... Thread.currentThread().sleep(1000); // Do something } catch(InterruptedException ie) { Thread.currentThread().interrupt(); // handleReset interrupted exceptionstatus } } } public synchronized boolean isDone() { return done; } public protectedsynchronized void shutdown() { done.set(true) = true; } } |
Although this compliant solution is acceptable, intrinsic locks cause threads to block and may introduce contention. On the other hand, volatile-qualified shared variables do not block. Excessive synchronization can also make the program prone to deadlock.
Synchronization is a more secure alternative in situations where the volatile
keyword or a java.util.concurrent.atomic.Atomic*
field is inappropriate, such as when a variable's new value depends on its current value (see VNA02-J. Ensure that compound operations on shared variables are atomic for more information).
Compliance with LCK00-J. Use private final lock objects to synchronize classes that may interact with untrusted code can reduce the likelihood of misuse by ensuring that untrusted callers cannot access the lock object.
Exceptions
VNA00-J-EX0: Class
objects are created by the virtual machine; their initialization always precedes any subsequent use. Consequently, cross-thread visibility of Class
objects is already assured by default.
Risk Assessment
Failing to ensure the visibility of a shared primitive variables on accesses can lead to variable may result in a thread seeing observing a stale values value of the variablesvariable.
Rule | Severity | Likelihood | Remediation Cost | Priority | Level |
---|
VNA00-J |
Medium |
Probable |
Medium | P8 | L2 |
Automated Detection
TODO
Related Vulnerabilities
Search for vulnerabilities resulting from the violation Some static analysis tools are capable of detecting violations of this rule on the CERT website.
References
Wiki Markup |
---|
\[[JLS 05|AA. Java References#JLS 05]\] [Chapter 17, Threads and Locks|http://java.sun.com/docs/books/jls/third_edition/html/memory.html], section 17.4.5 Happens-before Order, section 17.4.3 Programs and Program Order, section 17.4.8 Executions and Causality Requirements
\[[Tutorials 08|AA. Java References#Tutorials 08]\] [Java Concurrency Tutorial|http://java.sun.com/docs/books/tutorial/essential/concurrency/index.html]
\[[Lea 00|AA. Java References#Lea 00]\] Sections, 2.2.7 The Java Memory Model, 2.2.5 Deadlock, 2.1.1.1 Objects and locks
\[[Bloch 08|AA. Java References#Bloch 08]\] Item 66: Synchronize access to shared mutable data
\[[Goetz 06|AA. Java References#Goetz 06]\] 3.4.2. "Example: Using Volatile to Publish Immutable Objects"
\[[JPL 06|AA. Java References#JPL 06]\] 14.10.3. "The Happens-Before Relationship"
\[[MITRE 09|AA. Java References#MITRE 09]\] [CWE ID 667|http://cwe.mitre.org/data/definitions/667.html] "Insufficient Locking", [CWE ID 413|http://cwe.mitre.org/data/definitions/413.html] "Insufficient Resource Locking", [CWE ID 366|http://cwe.mitre.org/data/definitions/366.html] "Race Condition within a Thread", [CWE ID 567|http://cwe.mitre.org/data/definitions/567.html] "Unsynchronized Access to Shared Data" |
.
Tool | Version | Checker | Description | ||||||
---|---|---|---|---|---|---|---|---|---|
CodeSonar |
| JAVA.CONCURRENCY.LOCK.ICS | Impossible Client Side Locking (Java) | ||||||
Eclipse | 4.2.0 | Not Implemented | |||||||
FindBugs | 2.0.1 | Not Implemented | |||||||
Parasoft Jtest |
| CERT.VNA00.LORD CERT.VNA00.MRAV | Ensure that nested locks are ordered correctly Access related Atomic variables in a synchronized block | ||||||
PMD | 5.0.0 | Not Implemented | |||||||
Fortify | Not Implemented | ||||||||
Coverity | 7.5 | SERVLET_ATOMICITY | Implemented | ||||||
ThreadSafe |
| CCE_SL_INCONSISTENT | Implemented |
Related Guidelines
CWE-413, Improper Resource Locking |
Bibliography
Item 66, "Synchronize Access to Shared Mutable Data" | |
Section 3.4.2, "Example: Using Volatile to Publish Immutable Objects" | |
[JLS 2015] | Chapter 17, "Threads and Locks" |
[JPL 2006] | Section 14.10.3, "The Happens-Before Relationship" |
...
11. Concurrency (CON) 11. Concurrency (CON) CON02-J. Always synchronize on the appropriate object