flock tests: cover the case of asking for both locks on the same file handle

[RalfJung]

No description provided.

[rustbot]

Thank you for contributing to Miri! A reviewer will take a look at your PR, typically within a week or two.
Please remember to not force-push to the PR branch except when you need to rebase due to a conflict or when the reviewer asks you for it.

[rustbot]

Thank you for contributing to Miri! A reviewer will take a look at your PR, typically within a week or two.
Please remember to not force-push to the PR branch except when you need to rebase due to a conflict or when the reviewer asks you for it.

[RalfJung]

Strange, I would have expected this to work on Windows

    // Test that we can apply many shared locks
    file1.lock_shared().unwrap();
    file2.lock_shared().unwrap();
    // Test that shared lock prevents exclusive lock
    assert!(matches!(file1.try_lock().unwrap_err(), fs::TryLockError::WouldBlock));
    // Unlock one shared lock.
    file2.unlock().unwrap();
    // Now just file1 has a shared lock -- and we can convert that to an exclusive lock.
    file1.try_lock().unwrap(); // <-- fails here

But it seems like somehow releasing the lock on file2 does not actually make the lock available for file1? @ChrisDenton any idea what might be happening?

[ChrisDenton]

Oh right, sorry, it's just the exclusive lock then shared lock case that works. Doing the opposite doesn't.

A shared lock can overlap an exclusive lock if both locks were created using the same file handle.

[RalfJung]

Wow that's... creative. Thanks for clarifying.

[RalfJung]

FWIW I'd argue both locks were "created using the same file handle" in that example... they are both created via file1.