Page 323 - DCAP404 _Object Oriented Programming

Object-oriented Programming

Notes CallSomeMethod();
—SharedVar;
}
}
When this code is executed as a single thread, CallSomeMethod will execute 20 times. However,
imagine that we start a number of threads, all executing ThreadOne(). However, with more than
one thread existing simultaneously the method CallSomeMethod() will most likely be executed
too many times. Exactly how many times depends on the number of threads spawned, computer
architecture, operating system scheduling and lick. The problem arises because we do not test
and update SharedVar as an atomic operation, so there is a period where the value of SharedVar
is incorrect. During this time other threads can pass the test when they really shouldn’t have.

Did u know? What is the purpose of SharedVar?
The value of SharedVar on exit tells us how many extra times CallSomeMethod() is
called. The value of SharedVar on exit will be 0 if there is a single thread running
CallSomeMethod().
Race conditions can be avoided using an object called mutex.

A mutex is synchronization primitive provided by the operating system that can be used to
ensure that a section of code executes by one thread at a time. A mutex has two states - locked and
unlocked. In locked state any further attempt to lock it will suspend the thread. The waiting
threads can acquire lock on mutex and resume execution only when the mutex becomes unlocked.
The thread that locks the mutex can only unlocked it. Here is the solution for the above code
using mutex.
int SharedVar = 20;
mutex MutexVar;
void ThreadTwo()
{

while(SharedVar > 0)
{
bool flag = false;
{
mutex::lock(MutexVar);
if(SharedVar > 0)
{
—SharedVar;
flag = true;
}
}
if(flag) CallSomeMethod();
}
}

316 LOVELY PROFESSIONAL UNIVERSITY