we have a shared lock variable which can take one of the two values, 0 or 1

The Test_and_Set takes a shared variable called offset that represent the lock status and return its
original value and then set the variable to true which means the lock is disabled

in Test_and_Set entry section

in Test_and_set exit section

CompareAndSwap software instruction is also an atomic instruction

CompareAndSwap always returns the new value of the lock variable

in CompareAndSwap entry section

in CompareAndSwap exit section

Simplest is mutex lock, boolean variable indicating if lock is available or not

Calls to acquire () and release () can’t be atomi

in Mutex Locks Spinlocks is ideal for single-CPU as it is wasteful of CPU cycles that other process might be able to use productively

semaphore S is an integer variable that can only be modified via two interruptible operations: wait () and signal ()

in semaphore wait function the value of the integer is

in semaphore signal function the value of the integer is

integer value can range only between 0 and 1

a mutex lock

The main issue with semaphores is the busy loop in the signal() which eats CPU cycles without
accomplishing anything useful

semaphore implementation with no busy waiting there’s an associated waiting queue

remove one of processes in the waiting queue and place it in the ready queue