提升scoped_lock返回行为

Question

Is the two code segments produce same behavior? I think it does unless there is something I'm missing. Note: I have used this pointer everywhere just to clarify that all variables are member variables of some_class

int some_class::some_func()
{
    boost::scoped_lock lock(this->m_mutex);
    return this->member;
}

int some_class::some_func()
{
    this->m_mutex.lock();
    int a = this->member;
    this->m_mutex.unlock();
    return a;
}

Answer 1

some how these are doing same work but: this is not just about forgetting to call unlock() : an exception may occur while your mutex is locked, and your call to unlock() may never be reached, even though you do not have any return statement between your call to lock() and your call to unlock(); like:

m.lock() // m is a mutex
    // do something
    foo(); // If this throws an exception,that your mutex won't get unlocked
    // do something
    m.unlock()

Bur in the case of scoped_lock , the destructor of scoped_lock guard will be invoked during stack unwinding, making sure that associated mutex always gets released.

Answer 2

It's the same assuming that int a = this->member; doesn't throw an exception.