synchronization

Advanced Operating Systems (CS 5500)

Assignment 2 (120 points)

Due by 23rd March (Monday) 11:59pm

You are allowed to discuss the problem and solution design with others, but the code you submit

must be your own. Your solution must include the certification of authenticity “I certify that the

code in the method functions including method function main of this project are entirely my own

work.”

Process Synchronization: Producer-Consumer Problem

The purpose of this programming project is to explore process synchronization. This will be

accomplished by writing a program on the Producer / Consumer problem described below. Your

simulation will be implemented using Pthreads. This assignment is a modification to the

programming project “The Producer – Consumer Problem” found at the end of Chapter 7 of our

textbook. 1. Your program must be written using C or C++ and you are required to use the

Pthread with mutex and semaphore libraries.

In chapter 3, we discussed how a "bounded buffer" could be used to enable producer and

consumer processes to share memory. We described a technique using a circular buffer that can

hold BUFFER_SIZE-1 items. By using a shared memory location count, the buffer can hold all

BUFFER_SIZE items. This count is initialized to 0 and is incremented every time an item is

placed into the buffer and decremented every time an item is removed from the buffer. The

count data item can also be implemented as a counting semaphore.

The producer can place items into the buffer only if the buffer has a free memory location to

store the item. The producer cannot add items to a full buffer. The consumer can remove items

from the buffer if the buffer is not empty. The consumer must wait to consume items if the buffer

is empty.

The "items" stored in this buffer will be integers. Your producer process will have to insert

random numbers into the buffer. The consumer process will consume a number.

Assignment Specifications

The buffer used between producer and consumer processes will consist of a fixed-size array of

type buffer_item. The queue of buffer_item objects will be manipulated using a circular array.

The buffer will be manipulated with two functions, buffer_insert_item() and

buffer_remove_item(), which are called by the producer and consumer threads, respectively. A

skeleton outlining these functions can be found in buffer.h (provided with this assignment).

The buffer_insert_item() and buffer_remove_item() functions will synchronize the producer and

consumer using the algorithms. The buffer will also require an initialization function (not

supplied in buffer.h) that initializes the mutual exclusion object "mutex" along with the "empty"

and "full" semaphores.

The producer thread will alternate between sleeping for a random period of time and generating

and inserting (trying to) an integer into the buffer. Random numbers will be generated using the

rand_r() function. See the text on page 290 for an overview of the producer algorithm.

The consumer thread will alternate between sleeping for a random period of time (thread safe of

course) and (trying to) removing a number out of the buffer. See the text on page 290 for an

overview of the consumer algorithm.

The main function will initialize the buffer and create the separate producer and consumer

threads. Once it has created the producer and consumer threads, the main() function will sleep

for duration of the simulation. Upon awakening, the main thread will signal other threads to quit

by setting a simulation flag which is a global variable. The main thread will join with the other

threads and then display the simulation statistics. The main() function will be passed two

parameters on the command line:

• The length of time the main thread is to sleep before terminating (simulation length in

seconds)

• The maximum length of time the producer and consumer threads will sleep prior to

producing or consuming a buffer_item

A skeleton for the main function appears as:

#include <buffer.h>

int main( int argc, char *argv[] ){

Get command line arguments

Initialize buffer

Create producer thread(s)

Create consumer thread(s)

Sleep

Join Threads

Display Statistics

Exit

}

Creating Pthreads using the Pthreads API is discussed in Chapter 4 and in Assignment-1. Please

refer to those references for specific instructions regarding creation of the producer and

consumer Pthreads.

The following code sample illustrates how mutex locks available in the Pthread API can be used

to protect a critical section:

#include <pthread.h>

pthread_mutex_t mutex;

/* create the mutex lock */

pthread_mutex_init( &mutex, NULL );

/* aquire the mutex lock */

pthread_mutex_lock( &mutex );

/*** CRITICAL SECTION ***/

/* release the mutex lock */

pthread_mutex_unlock( &mutex );

Pthreads uses the pthread_mutex_t data type for mutex locks. A mutex is created with the

pthread_mutex_init() function, with the first parameter being a pointer to the mutex. By passing

NULL as a second parameter, we initialize the mutex to its default attributes. The mutex is

acquired and released with the pthread_mutex_lock() and pthread_mutex_unlock() functions. If

the mutex lock is unavailable when pthread_mutex_lock() is invoked, the calling thread is

blocked until the owner invokes pthread_mutex_unlock(). All mutex functions return a value of

0 with correct operation; if an error occurs, these functions return a nonzero error code.

Pthreads provides two types of semaphores: named and unnamed. For this project, we will use

unnamed semaphores. The code below illustrates how a semaphore is created:

#include <semaphore.h>

sem_t sem;

/* create the semaphore and initialize it to 5 */

sem_init( &sem, 0, 5 );

The sem_init() function creates and initializes a semaphore. This function is passed three

parameters: A pointer to the semaphore, a flag indicating the level of sharing, and the

semaphore's initial value. In this example, by passing the flag 0, we are indicating that this

semaphore can only be shared by threads belonging to the same process that created the

semaphore. A nonzero value would allow other processes to access the semaphore as well. In

this example, we initialize the semaphore to the value 5.

In Chapter-6 (Section 6.6), we described the classical wait() and signal() semaphore operations.

Pthread names the wait() and signal() operations sem_wait() and sem_post(), respectively. The

code example below creates a binary semaphore mutex with an initial value 1 and illustrates it

use in protecting a critical section:

#include <semaphore.h>

sem_t mutex;

/* create the semaphore */

sem_init( &mutex, 0, 1 );

/* acquire the semaphore */

sem_wait( &mutex );

/*** CRITICAL SECTION ***/

/* release the semaphore */

sem_post( &mutex );

Program Output

Your simulation should output when various conditions occur: buffer empty/full, location of

producer/consumer, etc.

Submission Guidelines and Requirements

1. Your program must be written using C or C++ and you are required to use the Pthread with mutex and semaphore libraries

2. You may use the C/C++ STL (Standard Template Library) in your solution. 3. You should use Netbeans to implement the assignment. You can download Netbeans

with C/C++ features from the following link: https://netbeans.org/downloads/8.2/

4. Create project in Netbeans for completing this assignment. 5. Include your name and UCM ID in the code: 6. Add the following statement in your solution “I certify that the code in the method

functions including method function main of this project are entirely my own work.”

// Your Name

// Your UCM ID

//Certificate of Authenticity: “I certify that the code in the method functions including

method function main of this project are entirely my own work.”

7. Add comments (about the function/variable/class) to your code as much as possible 8. Zip your project including source files and input/output text data files (if any) 9. Upload the zipped project file onto Blackboard