CODEWITHSUNDEEP
javaconcurrencythread-safetyfindbugs
Chris Vest
Chris Vest

Reputation: 8682

Possible FindBugs false positive of UL_UNRELEASED_LOCK_EXCEPTION_PATH?

I have code like this:

public class Thingy {
  private final Lock lock = new ReentrantLock();
  private boolean shutdown;
  public void shutdown() {
    lock.lock();
    shutdown = true;
    lock.unlock();
  }
}

And FindBugs complains that "Thingy.shutdown() does not release lock on all exception paths" and that I should wrap the shutdown = true; line in a try-finally, but as far as I can see there is no way that this can go bad.

Am I wrong or is this a false-positive?

Upvotes: 4

Views: 1941

Answers (2)

Jed Wesley-Smith
Jed Wesley-Smith

Reputation: 4706

Its true, but so is the fact that using a lock here is the wrong thing. You only need volatile.

In fact you don't even need that as you never read the value, so you can elide the variable and the lock altogether.

Point being that for a simple write, findbugs warning something is probably correct, it just is warning the wrong thing.

Upvotes: 2

dty
dty

Reputation: 18998

To be fair, it's generally fairly tricky (without being a full-on compiler and doing the analysis) to determine whether any given piece of code can throw an exception or not.

But I agree, in this case, it's a false positive. And even if it isn't - even if there's some dark corner of the JLS/JVM spec that says an assignment can, somehow, through an exception - if you get into that state, you've got more to worry about than an un-released lock!

Upvotes: 5

Related Questions