fork – My Programming Notes

Tag Archives: fork

C++ || Snippet – How To Use Message Queues For Interprocess Communication

October 04, 2013 admin No comments

The following is sample code which demonstrates the use of the msgsnd, msgrcv, and msgget function calls for use with message queues on Unix based systems.

Message queues are used for interprocess communication. They provide two (or more) separate programs (i.e: sender and receiver) with the ability to interact with eachother. This is ideal because the sender and receiver do not need to interact with the message queue at the same time. Messages can be sent at one point in time, be placed on the queue until the receiver is ready for it, and then be removed at another point in time when the program(s) that service the queue are finally ready to receive a message.

This concept is different from threaded and forked process IPC procedures, which often times process data in a more streamlined manner. Rather than following a strict pattern as to when data is to be sent and received, queuing is a mechanism in which messages are held until the receiving application is ready to process them.

The example on this page demonstrates the above use of a message queue to transfer data between two separate programs. The sending program (client.cpp) sends an integer and string variable to the receiving program (server.cpp), which then displays that received data to the screen.

NOTE: The server program must be ran before the client program!

=== 1. Server.cpp (Receiver) ===





		
		
			#1 - How To Use Message Queues - Receiver
			
C++
			
			// ============================================================================
//    Author: Kenneth Perkins
//    Date:   Oct 4, 2013
//    Taken From: http://programmingnotes.org/
//    File:  Server.cpp
//    Description: Demonstrate the use of a message queue for interprocess
//       communication. This program illustrates the server (receiving) end of
//       the message queue.
// ============================================================================
#include <iostream>
#include <cstdlib>
#include <unistd.h>
#include <sys/types.h>
#include <sys/ipc.h>
#include <sys/msg.h>
using namespace std;

// Compile & Run
// g++ Server.cpp -o Server
// ./Server

// the structure representing the message queue
// & must match the same layout as in the client.cpp
struct MsgQueue
{
    // IMPORTANT: every message structure must start with this
    long messageType;

    // these variables are optional & you can add 
    // more or less if you wish
    int someNumber;
    char buff[100];	

};

// message queue flag
const int MSG_Q_KEY_FLAG = 0664;

// message queue data transfer channel
const int MSG_Q_CHANNEL = 26;

int main()
{
    // declare variables
    key_t key = -1;
    int msqid = -1;
    MsgQueue msg;

    // use a random file and a random character to generate
    // a unique key. The same parameters to this function will 
    // always generate the same value. This is how multiple
    // processes can connect to the same queue.
    key = ftok("/bin/ls", 'K');

    // was the key allocation successful ?
    if(key < 0)
    {
        perror("ftok error");
        exit(1);
    }	

    // allocate the message queue if it does not already exist.
    // this function returns the id of the queue.
    msqid = msgget(key, MSG_Q_KEY_FLAG | IPC_CREAT);

    // was the allocation a success ?
    if(msqid < 0)
    {
        perror("msgget");
        exit(1);
    }

    // display info to the screen
    cout <<"\nThe server has started!\n"
        <<"\nWaiting for someone to connect to server id #"<<msqid<<" with "
        <<"the key "<<key<<endl<<endl;
                
    // recieve 10 messages from the client
    for(int x = 0; x < 10; ++x)
    {
        // this is where we receive messages:
        // @param: msqid - the id of the message queue
        // @param: msg - the message structure which stores the
        //   received message
        // @param: sizeof(msg) - sizeof(long) - size of the message
        //   excluding the required first member (messageType) which is
        //   required.
        // @param: MSG_Q_CHANNEL - receive all messages whose type parameter
        //   is set equal to "MSG_Q_CHANNEL"
        // @param: 0 - flag values (not useful for this example).
        if(msgrcv(msqid, &msg, sizeof(msg) - sizeof(long), MSG_Q_CHANNEL, 0) < 0)
        {
            perror("msgrcv");
            exit(1);
        }
        
        // print the received message from the client
        cout << "someNumber = "<<msg.someNumber<<" buff = "<<msg.buff<<endl;
    }

    // finally, deallocate the message queue
    if(msgctl(msqid, IPC_RMID, NULL) < 0)
    {
        perror("msgctl");
        exit(1);
    }
    
    cout << "\nServer is now shutting down!\n";

    return 0;
}// http://programmingnotes.org/
			
				
					
				
					1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
				
						// ============================================================================
//    Author: Kenneth Perkins
//    Date:   Oct 4, 2013
//    Taken From: http://programmingnotes.org/
//    File:  Server.cpp
//    Description: Demonstrate the use of a message queue for interprocess
//       communication. This program illustrates the server (receiving) end of
//       the message queue.
// ============================================================================
#include <iostream>
#include <cstdlib>
#include <unistd.h>
#include <sys/types.h>
#include <sys/ipc.h>
#include <sys/msg.h>
using namespace std;
 
// Compile & Run
// g++ Server.cpp -o Server
// ./Server
 
// the structure representing the message queue
// & must match the same layout as in the client.cpp
struct MsgQueue
{
    // IMPORTANT: every message structure must start with this
    long messageType;
 
    // these variables are optional & you can add 
    // more or less if you wish
    int someNumber;
    char buff[100];	
 
};
 
// message queue flag
const int MSG_Q_KEY_FLAG = 0664;
 
// message queue data transfer channel
const int MSG_Q_CHANNEL = 26;
 
int main()
{
    // declare variables
    key_t key = -1;
    int msqid = -1;
    MsgQueue msg;
 
    // use a random file and a random character to generate
    // a unique key. The same parameters to this function will 
    // always generate the same value. This is how multiple
    // processes can connect to the same queue.
    key = ftok("/bin/ls", 'K');
 
    // was the key allocation successful ?
    if(key < 0)
    {
        perror("ftok error");
        exit(1);
    }	
 
    // allocate the message queue if it does not already exist.
    // this function returns the id of the queue.
    msqid = msgget(key, MSG_Q_KEY_FLAG | IPC_CREAT);
 
    // was the allocation a success ?
    if(msqid < 0)
    {
        perror("msgget");
        exit(1);
    }
 
    // display info to the screen
    cout <<"\nThe server has started!\n"
        <<"\nWaiting for someone to connect to server id #"<<msqid<<" with "
        <<"the key "<<key<<endl<<endl;
                
    // recieve 10 messages from the client
    for(int x = 0; x < 10; ++x)
    {
        // this is where we receive messages:
        // @param: msqid - the id of the message queue
        // @param: msg - the message structure which stores the
        //   received message
        // @param: sizeof(msg) - sizeof(long) - size of the message
        //   excluding the required first member (messageType) which is
        //   required.
        // @param: MSG_Q_CHANNEL - receive all messages whose type parameter
        //   is set equal to "MSG_Q_CHANNEL"
        // @param: 0 - flag values (not useful for this example).
        if(msgrcv(msqid, &msg, sizeof(msg) - sizeof(long), MSG_Q_CHANNEL, 0) < 0)
        {
            perror("msgrcv");
            exit(1);
        }
        
        // print the received message from the client
        cout << "someNumber = "<<msg.someNumber<<" buff = "<<msg.buff<<endl;
    }
 
    // finally, deallocate the message queue
    if(msgctl(msqid, IPC_RMID, NULL) < 0)
    {
        perror("msgctl");
        exit(1);
    }
    
    cout << "\nServer is now shutting down!\n";
 
    return 0;
}// http://programmingnotes.org/

=== 2. Client.cpp (Sender) ===





		
		
			#2 - How To Use Message Queues - Sender
			
C++
			
			// ============================================================================
//    Author: Kenneth Perkins
//    Date:   Oct 4, 2013
//    Taken From: http://programmingnotes.org/
//    File:  Client.cpp
//    Description: Demonstrate the use of a message queue for interprocess
//       communication. This program illustrates the client (sending) end of
//       the message queue
// ============================================================================
#include <iostream>
#include <cstdlib>
#include <cstring>
#include <unistd.h>
#include <sys/types.h>
#include <sys/ipc.h>
#include <sys/msg.h>
using namespace std;

// Compile & Run
// g++ Client.cpp -o Client
// ./Client

// the structure representing the message queue
// & must match the same layout as in the server.cpp
struct MsgQueue
{
    // IMPORTANT: every message structure must start with this
    long messageType;

    // these variables are optional & you can add 
    // more or less if you wish
    int someNumber;
    char buff[100];	

};

// message queue flag
const int MSG_Q_KEY_FLAG = 0664;

// message queue type
const int MSG_Q_CHANNEL = 26;

int main()
{
    // declare variables
    key_t key = -1;
    int msqid = -1;
    MsgQueue msg;
        
    // use a random file and a random character to generate
    // a unique key. The same parameters to this function will 
    // always generate the same value. This is how multiple
    // processes can connect to the same queue.    
    key = ftok("/bin/ls", 'K');

    // was the key allocation successful ?
    if(key < 0)
    {
        perror("ftok");
        exit(1);
    }	

    // connect to the message queue; fail if the	
    // there is no message queue associated with
    // this key. This function returns the id of 
    // the queue.
    msqid = msgget(key, MSG_Q_KEY_FLAG);	

    // was the allocation a success ?
    if(msqid < 0)
    {
        perror("msgget");
        exit(1);	
    }	

    // display info to the screen
    cout <<"\nSuccessfully connected to server id #"<<msqid<<" with "
        <<"the key "<<key
        <<"\n\nNow sending messages....";
                
    // send 10 messages to the server
    for(int x = 0; x < 10; ++x)
    {
        // set the message type - this must match
        // the 4th parameter of msgrcv() in the server.cpp code
        msg.messageType = MSG_Q_CHANNEL;
        
        // place data into the message queue structure to send to the server
        msg.someNumber = x;
        strncpy(msg.buff, "Message queues are awesome!", sizeof(msg.buff));
        
        // this is where we send messages:
        // @param: msqid - the id of the message queue
        // @param: msg - the message structure which stores the
        //   message to send
        // @param: sizeof(msg) - sizeof(long) - size of the message
        //   excluding the required first member (messageType) which is
        //   required.
        // @param: 0 - flag values (not useful for this example).
        if(msgsnd(msqid, &msg, sizeof(msg) - sizeof(long), 0) < 0)
        {
            perror("msgsnd");
            exit(1);
        }
    }
    
    cout << "Sending complete!\n";

    return 0;
}// http://programmingnotes.org/
			
				
					
				
					1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
				
						// ============================================================================
//    Author: Kenneth Perkins
//    Date:   Oct 4, 2013
//    Taken From: http://programmingnotes.org/
//    File:  Client.cpp
//    Description: Demonstrate the use of a message queue for interprocess
//       communication. This program illustrates the client (sending) end of
//       the message queue
// ============================================================================
#include <iostream>
#include <cstdlib>
#include <cstring>
#include <unistd.h>
#include <sys/types.h>
#include <sys/ipc.h>
#include <sys/msg.h>
using namespace std;
 
// Compile & Run
// g++ Client.cpp -o Client
// ./Client
 
// the structure representing the message queue
// & must match the same layout as in the server.cpp
struct MsgQueue
{
    // IMPORTANT: every message structure must start with this
    long messageType;
 
    // these variables are optional & you can add 
    // more or less if you wish
    int someNumber;
    char buff[100];	
 
};
 
// message queue flag
const int MSG_Q_KEY_FLAG = 0664;
 
// message queue type
const int MSG_Q_CHANNEL = 26;
 
int main()
{
    // declare variables
    key_t key = -1;
    int msqid = -1;
    MsgQueue msg;
        
    // use a random file and a random character to generate
    // a unique key. The same parameters to this function will 
    // always generate the same value. This is how multiple
    // processes can connect to the same queue.    
    key = ftok("/bin/ls", 'K');
 
    // was the key allocation successful ?
    if(key < 0)
    {
        perror("ftok");
        exit(1);
    }	
 
    // connect to the message queue; fail if the	
    // there is no message queue associated with
    // this key. This function returns the id of 
    // the queue.
    msqid = msgget(key, MSG_Q_KEY_FLAG);	
 
    // was the allocation a success ?
    if(msqid < 0)
    {
        perror("msgget");
        exit(1);	
    }	
 
    // display info to the screen
    cout <<"\nSuccessfully connected to server id #"<<msqid<<" with "
        <<"the key "<<key
        <<"\n\nNow sending messages....";
                
    // send 10 messages to the server
    for(int x = 0; x < 10; ++x)
    {
        // set the message type - this must match
        // the 4th parameter of msgrcv() in the server.cpp code
        msg.messageType = MSG_Q_CHANNEL;
        
        // place data into the message queue structure to send to the server
        msg.someNumber = x;
        strncpy(msg.buff, "Message queues are awesome!", sizeof(msg.buff));
        
        // this is where we send messages:
        // @param: msqid - the id of the message queue
        // @param: msg - the message structure which stores the
        //   message to send
        // @param: sizeof(msg) - sizeof(long) - size of the message
        //   excluding the required first member (messageType) which is
        //   required.
        // @param: 0 - flag values (not useful for this example).
        if(msgsnd(msqid, &msg, sizeof(msg) - sizeof(long), 0) < 0)
        {
            perror("msgsnd");
            exit(1);
        }
    }
    
    cout << "Sending complete!\n";
 
    return 0;
}// http://programmingnotes.org/

QUICK NOTES:
The highlighted lines are sections of interest to look out for.

You can view all allocated message queues using the ipcs command. You can delete a message queue from command line using ipcrm -q [KEY SHOWN BY IPCS]. Message queues are a finite resource. If something goes wrong during the execution of your program, you must manually delete all your queues.

The code is heavily commented, so no further insight is necessary. If you have any questions, feel free to leave a comment below.

The following is sample output.

SERVER OUTPUT:

The server has started!
Waiting for someone to connect to server id #753664 with the key 1258295474 someNumber = 0 buff = Message queues are awesome! someNumber = 1 buff = Message queues are awesome! someNumber = 2 buff = Message queues are awesome! someNumber = 3 buff = Message queues are awesome! someNumber = 4 buff = Message queues are awesome! someNumber = 5 buff = Message queues are awesome! someNumber = 6 buff = Message queues are awesome! someNumber = 7 buff = Message queues are awesome! someNumber = 8 buff = Message queues are awesome! someNumber = 9 buff = Message queues are awesome!
Server is now shutting down!

CLIENT OUTPUT:

Successfully connected to server id #753664 with the key 1258295474
Now sending messages....Sending complete!

C++ || Snippet – How To Create And Use Threads For Interprocess Communication

October 03, 2013 admin No comments

The following is sample code which demonstrates the use of POSIX threads (pthreads), aswell as the pthread_create, and pthread_join function calls on Unix based systems.

Much like the fork() function call which is used to create new processes, threads are similar in that they too are used for interprocess communication. Threads allow for multi-threading, which is a widespread programming and execution model that allows for multiple threads to exist within the same context of a single program.

Threads share the calling parent process’ resources. Each process has it’s own address space, but the threads within the same process share that address space. Threads also share any other resources within that process. This means that it’s very easy to share data amongst threads, but it’s also easy for the threads to step on each other, which can lead to bad things.

The example on this page demonstrates the use of multiple pthreads to display shared data (an integer variable) to the screen.





		
		
			How To Use Multiple Pthreads
			
C++
			
			// ============================================================================
//    Author: Kenneth Perkins
//    Date:   Oct 3, 2013
//    Taken From: http://programmingnotes.org/
//    File:  Pthread.cpp
//    Description: Demonstrate the use of using multiple threads using the 
//       pthread_create() function
// ============================================================================
#include <iostream>
#include <cstdlib>
#include <pthread.h>
using namespace std;

// Compile & Run
// g++ Pthread.cpp -lpthread -o Pthread
// ./Pthread <NUMBER OF THREADS TO CREATE>

// function  prototype
void* ThreadFunc(void* args);

// global variable
int threadNum = 0;

int main(int argc, char* argv[])
{
    // declare variables
    int numberOfThreads = 0;
    pthread_t* threads;

    // make sure theres enough commandline arguments
    // and EXIT if theres not
    if(argc < 2)
    {
        cout <<"** ERROR - NOT ENOUGH ARGS!\n"
            <<"\nUSAGE:"<<argv[0]<<" <NUMBER OF THREADS TO CREATE>\n";
        exit(1);
    }

    // convert the number of threads from the command line
    // from string to int value
    numberOfThreads = atoi(argv[1]);

    // declare the array of thread variables 
    threads = new pthread_t[numberOfThreads];
    
    cout <<"\nThe Parent is creating "<<numberOfThreads<<" threads!" << endl;

    // use a loop to create the threads
    for(int x = 0; x < numberOfThreads; ++x)
    {
        // create the thread and call the "ThreadFunc"
        if(pthread_create(&threads[x], NULL, ThreadFunc, NULL) < 0)
        {
            perror("pthread_create error");
            exit(1);
        }
    }
    
    cout <<"The Parent is now waiting for the thread(s) to complete...\n\n";

    // wait for all threads to finish
    for(int x = 0; x < numberOfThreads; ++x)
    {
        pthread_join(threads[x], NULL);
    }
    
    cout <<"\nAll thread(s) are complete and have terminated!\n"
        <<"\nThe Parent is now exiting...\n";

    return 0;
}// end of main

/**
 * The function called by the thread
 * @param args - pointer to the thread local data
 */
void* ThreadFunc(void* args)
{	
    // display the thread info
    cout <<"Hi, Im thread #"<<++threadNum
        <<" and this is my id number: "<<(unsigned)pthread_self()<<endl;
    return 0;
}// http://programmingnotes.org/
			
				
					
				
					1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
				
						// ============================================================================
//    Author: Kenneth Perkins
//    Date:   Oct 3, 2013
//    Taken From: http://programmingnotes.org/
//    File:  Pthread.cpp
//    Description: Demonstrate the use of using multiple threads using the 
//       pthread_create() function
// ============================================================================
#include <iostream>
#include <cstdlib>
#include <pthread.h>
using namespace std;
 
// Compile & Run
// g++ Pthread.cpp -lpthread -o Pthread
// ./Pthread <NUMBER OF THREADS TO CREATE>
 
// function  prototype
void* ThreadFunc(void* args);
 
// global variable
int threadNum = 0;
 
int main(int argc, char* argv[])
{
    // declare variables
    int numberOfThreads = 0;
    pthread_t* threads;
 
    // make sure theres enough commandline arguments
    // and EXIT if theres not
    if(argc < 2)
    {
        cout <<"** ERROR - NOT ENOUGH ARGS!\n"
            <<"\nUSAGE:"<<argv[0]<<" <NUMBER OF THREADS TO CREATE>\n";
        exit(1);
    }
 
    // convert the number of threads from the command line
    // from string to int value
    numberOfThreads = atoi(argv[1]);
 
    // declare the array of thread variables 
    threads = new pthread_t[numberOfThreads];
    
    cout <<"\nThe Parent is creating "<<numberOfThreads<<" threads!" << endl;
 
    // use a loop to create the threads
    for(int x = 0; x < numberOfThreads; ++x)
    {
        // create the thread and call the "ThreadFunc"
        if(pthread_create(&threads[x], NULL, ThreadFunc, NULL) < 0)
        {
            perror("pthread_create error");
            exit(1);
        }
    }
    
    cout <<"The Parent is now waiting for the thread(s) to complete...\n\n";
 
    // wait for all threads to finish
    for(int x = 0; x < numberOfThreads; ++x)
    {
        pthread_join(threads[x], NULL);
    }
    
    cout <<"\nAll thread(s) are complete and have terminated!\n"
        <<"\nThe Parent is now exiting...\n";
 
    return 0;
}// end of main
 
/**
 * The function called by the thread
 * @param args - pointer to the thread local data
 */
void* ThreadFunc(void* args)
{	
    // display the thread info
    cout <<"Hi, Im thread #"<<++threadNum
        <<" and this is my id number: "<<(unsigned)pthread_self()<<endl;
    return 0;
}// http://programmingnotes.org/

QUICK NOTES:
The highlighted lines are sections of interest to look out for.

So, what is the difference between fork() processes and thread processes? Threads differ from traditional multitasking operating system processes in that:

• processes are typically independent, while threads exist as subsets of a process • processes carry considerable state information, whereas multiple threads within a process share state as well as memory and other resources • processes have separate address spaces, whereas threads share their address space • processes interact only through system-provided inter-process communication mechanisms. • Context switching between threads in the same process is typically faster than context switching between processes.

To use pthreads, the compiler flag “-lpthread” must be set.

The code is heavily commented, so no further insight is necessary. If you have any questions, feel free to leave a comment below.

The following is sample output:

./Pthread 5
The Parent is creating 5 threads! The Parent is now waiting for the thread(s) to complete... Hi, Im thread #1 and this is my id number: 1664468736 Hi, Im thread #2 and this is my id number: 1647683328 Hi, Im thread #3 and this is my id number: 1656076032 Hi, Im thread #4 and this is my id number: 1672861440 Hi, Im thread #5 and this is my id number: 1681254144 All thread(s) are complete and have terminated!
The Parent is now exiting...

C++ || Multi-Hash Interprocess Communication Using Fork, Popen, & Pipes

August 20, 2013 admin No comments

The following is another homework assignment which was presented in an Operating Systems Concepts class. Using two pipes, the following is a program which implements the computing of hash values on a file using the MD5, SHA1, SHA224, SHA256, SHA384, and SHA512 hashing algorithms provided on Unix based systems.

REQUIRED KNOWLEDGE FOR THIS PROGRAM

How To Use Fork How To Use Pipe How To Use Popen

==== 1. OVERVIEW ====

Hash algorithms map large data sets of variable length (e.g. files), to data sets of a fixed length. For example, the contents of a 1GB file may be hashed into a single 128-bit integer. Many hash algorithms exhibit an important property called an avalanche effect – slight changes in the input data trigger significant changes in the hash value.

Hash algorithms are often used for verifying the integrity of files downloaded from the WEB. For example, websites hosting a file usually post the hash value of the file using the MD5 hash algorithm. By doing this, the user can then verify the integrity of the downloaded file by computing the MD5 algorithm on their own, and compare their hash value against the hash value posted on the website. The user will know if the download was valid only if the two hash values match.

==== 2. TECHNICAL DETAILS ====

The following implements a program for computing the hash value of a file using the MD5, SHA1, SHA224, SHA256, SHA384, and SHA512 hashing algorithms provided on Unix based systems.

This program takes the name of the target file being analyzed as a command line argument, and does the following:

1. Check to make sure the file exists. 2. Create two pipes. 3. Create a child process. 4. The parent transmits the name of the file to the child (over the first pipe). 5. The child receives the name of the file and computes the hash of the file using the MD5 algorithm (using Linux program md5sum). 6. The child transmits the computed hash to the parent (over the second pipe) and terminates. 7. The parent receives the hash, prints it, and calls wait(). 8. Repeat the same process starting with step 3, but using algorithms SHA1...SHA512. 9. The parent terminates after all hashes have been computed.

The use of the popen function is used in order to launch the above programs and capture their output into a character array buffer.

This program also uses two pipes. The two pipes created are the following:

(1) Parent to child pipe: Used by the parent to transfer the name of the file to the child. The parent writes to this pipe and the child reads it.
(2) Child to parent pipe: Used by the child to transfer the computed hashes to the parent. The child writes to this pipe and the parent reads it.





		
		
			Multi-Hash Using Fork, Popen, & Pipes
			
C++
			
			// =============================================================================
//    Author: K Perkins
//    Date:   Aug 20, 2013
//    Taken From: http://programmingnotes.org/
//    File: Multihash.cpp
//    Description: Hash algorithms map large data sets of variable length (e.g. 
//       files), to data sets of a fixed length. For example, contents of a 
//       1GB file may be hashed into a single 128-bit integer. Using 2 pipes, 
//       this program implements the computing of hash values on a file using 
//       the MD5, SHA1, SHA224, SHA256, SHA384, and SHA512 hashing algorithms.
//
//                   The two pipes created are the following:
//       (1) Parent to child pipe: Used by the parent to transfer the name of 
//             the file to the child. The parent writes to this pipe and the 
//             child reads it.
//       (2) Child to parent pipe: Used by the child to transfer the computed 
//             hashes to the parent. The child writes to this pipe and the 
//             parent reads it.
// =============================================================================
#include <iostream>
#include <fstream>
#include <cstdlib>
#include <cstring>
#include <unistd.h>
#include <sys/wait.h>
using namespace std;

// compile: g++ Multihash.cpp -o Multihash
// run: ./Multihash <file name>

// parentToChild = pipe for parent to child communication
// childToParent = pipe for child to parent communication
int parentToChild[2];
int childToParent[2];

// names of the hash programs
const char hashAlgs[6][10] = {"md5sum", "sha1sum", "sha224sum", 
            "sha256sum", "sha384sum", "sha512sum"};

// read end of pipe
const int READ_END = 0;

// write end of pipe 
const int WRITE_END = 1;

// number of hash programs 
const int NUM_HASH_ALGS = sizeof(hashAlgs)/sizeof(hashAlgs[0]);

// maximum length of hash value 
const int HASH_LENGTH = 1000;

// maximum length of file name 
const int FILENAME_LENGTH = 100;

// child function which computes the hash of a file and 
// returns the value back to the parent
// @param 'hashAlg' == the name of the hash program
void ComputeHash(const char hashAlg[]);

int main(int argc, char* argv[])
{  
    // declare variables
    ifstream infile;  // used to see if file exists 	
    char hashValue[HASH_LENGTH]; // hash value which child passes to parent   

    // check to see if theres enough commandline args 
    if(argc < 2)
    {
        cerr<<"n** ERROR - NOT ENOUGH ARGUMENTS!n"
            <<"nUSAGE: "<<argv[0]<<" <FILE NAME>n";
        exit(1);
    } 

    // try to open file
    infile.open(argv[1]);
   
    // if file doesnt exist, then exit
    if(infile.fail())
    {
        cerr<<"n** ERROR!n"        
            <<"Cant find the file ""<<argv[1]<<""!nn";
        exit(1);
    }
    infile.close();
    
    // start the piping process
    for(int currentAlg = 0; currentAlg < NUM_HASH_ALGS; ++currentAlg)
    {
        // create pipes
        if((pipe(parentToChild) < 0) || (pipe(childToParent) < 0))
        {
            cerr << "npipe failedn";
            exit(1);
        }	
        // create & fork a child
        pid_t pid = fork();

        // make sure fork succeeded 
        if(pid < 0)
        {
            cerr << "nfork failedn";
            exit(1);
        }		
        // child process	     
        else if(pid == 0) 
        { 
            ComputeHash(hashAlgs[currentAlg]);
        }	
        // parent process 
        else 
        {  
            // close selected pipes ends        	    	    
            close(parentToChild[READ_END]);
            close(childToParent[WRITE_END]);
            
            // pass the filename to the child	    
            write(parentToChild[WRITE_END], argv[1], strlen(argv[1])+1);
            close(parentToChild[WRITE_END]);
            
            // read the incoming hash value from the child
            read(childToParent[READ_END], hashValue, sizeof(hashValue));
            close(childToParent[READ_END]);
            
            // wait for child to complete
            wait(NULL);
            
            // display current hash to the screen
            cout<<"Hash Algorithm #"<<currentAlg+1<<":n"<<hashAlgs[currentAlg]
                    <<" - HASH VALUE: "<<hashValue<<endl;
                    
            // reset hash buffer
            memset(hashValue, (char)NULL, sizeof(hashValue));		            	     
        }  
    }	

    cerr<<"The parent process is now exiting...n";		
    return 0;	  
}// end of main

void ComputeHash(const char hashAlg[])
{
    // declare variables
    char recievedFileName[FILENAME_LENGTH]; // saves recieved filename
    char cmdLine[FILENAME_LENGTH]; // saves command line for popen
    char hashOutput[HASH_LENGTH]; // saves final hash output
    FILE* popenOutput; // popen file pointer
    
    // close selected pipes ends
    close(parentToChild[WRITE_END]);
    close(childToParent[READ_END]);

    // get filename from parent
    read(parentToChild[READ_END], recievedFileName, sizeof(recievedFileName));
    close(parentToChild[READ_END]);    
    
    // construct command line argument to pass to popen
    strncpy(cmdLine, hashAlg, sizeof(cmdLine));
    strncat(cmdLine, " ", sizeof(cmdLine));
    strncat(cmdLine, recievedFileName, sizeof(cmdLine));
    
    // get has value for current hash prog
    popenOutput = popen(cmdLine, "r");
    
    // make sure that popen succeeded 
    if(!popenOutput)
    {
        cerr<<"npopen failedn";
        exit(1);
    } 

    // set hash buffer to all NULLS 
    memset(hashOutput, (char)NULL, sizeof(hashOutput));	

    // read program output into buffer
    fread(hashOutput, sizeof(char), sizeof(char)*sizeof(hashOutput), popenOutput);

    // close popen buffer
    pclose(popenOutput);

    // pass hash value back to parent
    write(childToParent[WRITE_END], hashOutput, strlen(hashOutput)+1);
    close(childToParent[WRITE_END]); 
    
    // exit 'AKA' kill child   
    exit(0);
}// http://programmingnotes.org/
			
				
					
				
					1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
				
						// =============================================================================
//    Author: K Perkins
//    Date:   Aug 20, 2013
//    Taken From: http://programmingnotes.org/
//    File: Multihash.cpp
//    Description: Hash algorithms map large data sets of variable length (e.g. 
//       files), to data sets of a fixed length. For example, contents of a 
//       1GB file may be hashed into a single 128-bit integer. Using 2 pipes, 
//       this program implements the computing of hash values on a file using 
//       the MD5, SHA1, SHA224, SHA256, SHA384, and SHA512 hashing algorithms.
//
//                   The two pipes created are the following:
//       (1) Parent to child pipe: Used by the parent to transfer the name of 
//             the file to the child. The parent writes to this pipe and the 
//             child reads it.
//       (2) Child to parent pipe: Used by the child to transfer the computed 
//             hashes to the parent. The child writes to this pipe and the 
//             parent reads it.
// =============================================================================
#include <iostream>
#include <fstream>
#include <cstdlib>
#include <cstring>
#include <unistd.h>
#include <sys/wait.h>
using namespace std;
 
// compile: g++ Multihash.cpp -o Multihash
// run: ./Multihash <file name>
 
// parentToChild = pipe for parent to child communication
// childToParent = pipe for child to parent communication
int parentToChild[2];
int childToParent[2];
 
// names of the hash programs
const char hashAlgs[6][10] = {"md5sum", "sha1sum", "sha224sum", 
            "sha256sum", "sha384sum", "sha512sum"};
 
// read end of pipe
const int READ_END = 0;
 
// write end of pipe 
const int WRITE_END = 1;
 
// number of hash programs 
const int NUM_HASH_ALGS = sizeof(hashAlgs)/sizeof(hashAlgs[0]);
 
// maximum length of hash value 
const int HASH_LENGTH = 1000;
 
// maximum length of file name 
const int FILENAME_LENGTH = 100;
 
// child function which computes the hash of a file and 
// returns the value back to the parent
// @param 'hashAlg' == the name of the hash program
void ComputeHash(const char hashAlg[]);
 
int main(int argc, char* argv[])
{  
    // declare variables
    ifstream infile;  // used to see if file exists 	
    char hashValue[HASH_LENGTH]; // hash value which child passes to parent   
 
    // check to see if theres enough commandline args 
    if(argc < 2)
    {
        cerr<<"n** ERROR - NOT ENOUGH ARGUMENTS!n"
            <<"nUSAGE: "<<argv[0]<<" <FILE NAME>n";
        exit(1);
    } 
 
    // try to open file
    infile.open(argv[1]);
   
    // if file doesnt exist, then exit
    if(infile.fail())
    {
        cerr<<"n** ERROR!n"        
            <<"Cant find the file ""<<argv[1]<<""!nn";
        exit(1);
    }
    infile.close();
    
    // start the piping process
    for(int currentAlg = 0; currentAlg < NUM_HASH_ALGS; ++currentAlg)
    {
        // create pipes
        if((pipe(parentToChild) < 0) || (pipe(childToParent) < 0))
        {
            cerr << "npipe failedn";
            exit(1);
        }	
        // create & fork a child
        pid_t pid = fork();
 
        // make sure fork succeeded 
        if(pid < 0)
        {
            cerr << "nfork failedn";
            exit(1);
        }		
        // child process	     
        else if(pid == 0) 
        { 
            ComputeHash(hashAlgs[currentAlg]);
        }	
        // parent process 
        else 
        {  
            // close selected pipes ends        	    	    
            close(parentToChild[READ_END]);
            close(childToParent[WRITE_END]);
            
            // pass the filename to the child	    
            write(parentToChild[WRITE_END], argv[1], strlen(argv[1])+1);
            close(parentToChild[WRITE_END]);
            
            // read the incoming hash value from the child
            read(childToParent[READ_END], hashValue, sizeof(hashValue));
            close(childToParent[READ_END]);
            
            // wait for child to complete
            wait(NULL);
            
            // display current hash to the screen
            cout<<"Hash Algorithm #"<<currentAlg+1<<":n"<<hashAlgs[currentAlg]
                    <<" - HASH VALUE: "<<hashValue<<endl;
                    
            // reset hash buffer
            memset(hashValue, (char)NULL, sizeof(hashValue));		            	     
        }  
    }	
 
    cerr<<"The parent process is now exiting...n";		
    return 0;	  
}// end of main
 
void ComputeHash(const char hashAlg[])
{
    // declare variables
    char recievedFileName[FILENAME_LENGTH]; // saves recieved filename
    char cmdLine[FILENAME_LENGTH]; // saves command line for popen
    char hashOutput[HASH_LENGTH]; // saves final hash output
    FILE* popenOutput; // popen file pointer
    
    // close selected pipes ends
    close(parentToChild[WRITE_END]);
    close(childToParent[READ_END]);
 
    // get filename from parent
    read(parentToChild[READ_END], recievedFileName, sizeof(recievedFileName));
    close(parentToChild[READ_END]);    
    
    // construct command line argument to pass to popen
    strncpy(cmdLine, hashAlg, sizeof(cmdLine));
    strncat(cmdLine, " ", sizeof(cmdLine));
    strncat(cmdLine, recievedFileName, sizeof(cmdLine));
    
    // get has value for current hash prog
    popenOutput = popen(cmdLine, "r");
    
    // make sure that popen succeeded 
    if(!popenOutput)
    {
        cerr<<"npopen failedn";
        exit(1);
    } 
 
    // set hash buffer to all NULLS 
    memset(hashOutput, (char)NULL, sizeof(hashOutput));	
 
    // read program output into buffer
    fread(hashOutput, sizeof(char), sizeof(char)*sizeof(hashOutput), popenOutput);
 
    // close popen buffer
    pclose(popenOutput);
 
    // pass hash value back to parent
    write(childToParent[WRITE_END], hashOutput, strlen(hashOutput)+1);
    close(childToParent[WRITE_END]); 
    
    // exit 'AKA' kill child   
    exit(0);
}// http://programmingnotes.org/

QUICK NOTES:
The highlighted lines are sections of interest to look out for.

The code is heavily commented, so no further insight is necessary. If you have any questions, feel free to leave a comment below.

Using the following example input file located here, the following is sample output:





		
		
			Hash Algorithm #1:
md5sum - HASH VALUE: cd2f02cc5e50bd08d83cef630a32d7d6  INPUT_Dictionary_programmingnotes_freeweq_com.txt

Hash Algorithm #2:
sha1sum - HASH VALUE: 1e1d2fd77b331fc2b2f24822fdf2133f3678d662  INPUT_Dictionary_programmingnotes_freeweq_com.txt

Hash Algorithm #3:
sha224sum - HASH VALUE: bf69da0f9f9990cece8dd6800a25a9faca5381c0c62e0667c258e8d5  INPUT_Dictionary_programmingnotes_freeweq_com.txt

Hash Algorithm #4:
sha256sum - HASH VALUE: a8cd7a082ce571f7e66bc4d5eea8f71e7e455735922c90c060580b51d43593ec  INPUT_Dictionary_programmingnotes_freeweq_com.txt

Hash Algorithm #5:
sha384sum - HASH VALUE: a564df501470a43eebb51c28f83ab07090e39c434fc40bf496622f536c525140301633d6513dec83ea512bcb38c4e2e6  INPUT_Dictionary_programmingnotes_freeweq_com.txt

Hash Algorithm #6:
sha512sum - HASH VALUE: e686a709b3025f8b21fd6d40ce92f741dd8644335e935baced86098ff3e7278b5443e28f02fa886e7c0cd391ce6e8aa91842a6de29535a40453d41862a0bc1c7  INPUT_Dictionary_programmingnotes_freeweq_com.txt

The parent process is now exiting...
			
				
					
				
					1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
				
						Hash Algorithm #1:
md5sum - HASH VALUE: cd2f02cc5e50bd08d83cef630a32d7d6  INPUT_Dictionary_programmingnotes_freeweq_com.txt
 
Hash Algorithm #2:
sha1sum - HASH VALUE: 1e1d2fd77b331fc2b2f24822fdf2133f3678d662  INPUT_Dictionary_programmingnotes_freeweq_com.txt
 
Hash Algorithm #3:
sha224sum - HASH VALUE: bf69da0f9f9990cece8dd6800a25a9faca5381c0c62e0667c258e8d5  INPUT_Dictionary_programmingnotes_freeweq_com.txt
 
Hash Algorithm #4:
sha256sum - HASH VALUE: a8cd7a082ce571f7e66bc4d5eea8f71e7e455735922c90c060580b51d43593ec  INPUT_Dictionary_programmingnotes_freeweq_com.txt
 
Hash Algorithm #5:
sha384sum - HASH VALUE: a564df501470a43eebb51c28f83ab07090e39c434fc40bf496622f536c525140301633d6513dec83ea512bcb38c4e2e6  INPUT_Dictionary_programmingnotes_freeweq_com.txt
 
Hash Algorithm #6:
sha512sum - HASH VALUE: e686a709b3025f8b21fd6d40ce92f741dd8644335e935baced86098ff3e7278b5443e28f02fa886e7c0cd391ce6e8aa91842a6de29535a40453d41862a0bc1c7  INPUT_Dictionary_programmingnotes_freeweq_com.txt
 
The parent process is now exiting...

C++ || Serial & Parallel Multi Process File Downloader Using Fork & Execlp

August 19, 2013 admin 3 Comments

The following is another homework assignment which was presented in an Operating Systems Concepts class. The following are two multi-process programs using commandline arguments, which demonstrates more practice using the fork() and execlp() system calls on Unix based systems.

==== 1. OVERVIEW ====

File downloaders are programs used for downloading ﬁles from the Internet. The following programs listed on this page implement two distinct type of multi-process downloaders:

1. a serial ﬁle downloader which downloads ﬁles one by one. 2. a parallel ﬁle downloader which dowloads multiple ﬁles in parallel.

In both programs, the parent process first reads a ﬁle via the commandline. This file which is read is the ﬁle that contains the list of URLs of the ﬁles to be downloaded. The incoming url file that is read has the following format:

[URL1] [URL2] . . . [URLN]

Where [URL] is an http internet link with a valid absolute file path extension.
(i.e: http://newsimg.ngfiles.com/270000/270173_0204618900-cc-asmbash.jpg)

After the url file is parsed, next the parent process forks a child process. Each created child process uses the execlp() system call to replace its executable image with that of the “wget” program. The use of the wget program performs the actual file downloading.

==== 2. SERIAL DOWNLOADER ====

The serial downloader downloads files one at a time. After the parent process has read and parsed the incoming url file from the commandline, the serial downloader proceeds as follows:

1. The parent forks off a child process. 2. The child uses execlp("/usr/bin/wget", "wget", [URL STRING1], NULL) system call in order to replace its program with wget program that will download the first file in urls.txt (i.e. the file at URL ). 3. The parent executes a wait() system call until the child exits. 4. The parent forks off another child which downloads the next file specified in url.txt. 5. Repeat the same process until all files are downloaded.

The following is implemented below:





		
		
			Serial File Downloader Using fork() & execlp('wget')
			
C++
			
			// ============================================================================
//    Author: Kenneth Perkins
//    Date:   Aug 19, 2013
//    Taken From: http://programmingnotes.org/
//    File:  Serial.cpp
//    Description: File downloaders are programs used for downloading files
//       from the Internet. Using the fork() & execlp("wget") command, the
//       following is a multi-process serial file downloader which reads an
//       input file containing url file download links as a commandline
//       argument and downloads the files located on the internet one by one.
// ============================================================================
#include <iostream>
#include <cstring>
#include <fstream>
#include <cstdlib>
#include <unistd.h>
#include <sys/wait.h>
using namespace std;

// compile & run
// g++ Serial.cpp -o Serial
// ./Serial urls.txt

int main(int argc, char* argv[])
{
    // declare variables
    pid_t pid = -1;
    int urlNumber = 0;
    char urlName[256];
    ifstream infile;

    // check if theres enough command line args
    if(argc < 2)
    {
        cout <<"\nERROR -- NOT ENOUGH ARGS!"
            <<"\n\nUSAGE: "<<argv[0]<<" <file containing url downloads>\n\n";
        exit(1);
    }

    // try to open the file containing the download url links
    // exit if the url file is not found
    infile.open(argv[1]);
    if(infile.fail())
    {
        cout <<"\nERROR -- "<<argv[1]<<" NOT FOUND!\n\n";
        exit(1);
    }

    // get download url links from the file
    while(infile.getline(urlName, sizeof(urlName)))
    {
        ++urlNumber;

        // fork another process
        pid = fork();

        if(pid < 0)
        {   // ** error occurred
            perror("fork");
            exit(1);
        }
        else if(pid == 0)
        {   // ** child process
            cout <<endl<<"** URL #"<<urlNumber
                    <<" is currently downloading... **\n\n";
            execlp("/usr/bin/wget", "wget", urlName, NULL);
        }
        else
        {   // ** parent process
            // parent will wait for the child to complete
            wait(NULL);
            cout <<endl<<"-- URL #"<<urlNumber<<" is complete! --\n";
        }
    }
    infile.close();

    cout <<endl<<"The parent process is now exiting...\n";

    return 0;
}// http://programmingnotes.org/
			
				
					
				
					1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
				
						// ============================================================================
//    Author: Kenneth Perkins
//    Date:   Aug 19, 2013
//    Taken From: http://programmingnotes.org/
//    File:  Serial.cpp
//    Description: File downloaders are programs used for downloading files
//       from the Internet. Using the fork() & execlp("wget") command, the
//       following is a multi-process serial file downloader which reads an
//       input file containing url file download links as a commandline
//       argument and downloads the files located on the internet one by one.
// ============================================================================
#include <iostream>
#include <cstring>
#include <fstream>
#include <cstdlib>
#include <unistd.h>
#include <sys/wait.h>
using namespace std;
 
// compile & run
// g++ Serial.cpp -o Serial
// ./Serial urls.txt
 
int main(int argc, char* argv[])
{
    // declare variables
    pid_t pid = -1;
    int urlNumber = 0;
    char urlName[256];
    ifstream infile;
 
    // check if theres enough command line args
    if(argc < 2)
    {
        cout <<"\nERROR -- NOT ENOUGH ARGS!"
            <<"\n\nUSAGE: "<<argv[0]<<" <file containing url downloads>\n\n";
        exit(1);
    }
 
    // try to open the file containing the download url links
    // exit if the url file is not found
    infile.open(argv[1]);
    if(infile.fail())
    {
        cout <<"\nERROR -- "<<argv[1]<<" NOT FOUND!\n\n";
        exit(1);
    }
 
    // get download url links from the file
    while(infile.getline(urlName, sizeof(urlName)))
    {
        ++urlNumber;
 
        // fork another process
        pid = fork();
 
        if(pid < 0)
        {   // ** error occurred
            perror("fork");
            exit(1);
        }
        else if(pid == 0)
        {   // ** child process
            cout <<endl<<"** URL #"<<urlNumber
                    <<" is currently downloading... **\n\n";
            execlp("/usr/bin/wget", "wget", urlName, NULL);
        }
        else
        {   // ** parent process
            // parent will wait for the child to complete
            wait(NULL);
            cout <<endl<<"-- URL #"<<urlNumber<<" is complete! --\n";
        }
    }
    infile.close();
 
    cout <<endl<<"The parent process is now exiting...\n";
 
    return 0;
}// http://programmingnotes.org/

Since the serial downloader downloads files one at a time, that can become very slow. That is where the parallel downloader comes in handy!

==== 3. PARALLEL DOWNLOADER ====

The parallel downloader downloads files all at once and is implemented much like the serial downloader. The parallel downloader proceeds as follows:

1. The parent forks off n children, where n is the number of URLs in url.txt. 2. Each child executes execlp("/usr/bin/wget", "wget", [URL STRING], NULL) system call where each is a distinct URL in url.txt. 3. The parent calls wait() (n times in a row) and waits for all children to terminate. 4. The parent exits.

The following is implemented below:





		
		
			Parallel File Downloader Using fork() & execlp('wget')
			
C++
			
			// ============================================================================
//    Author: Kenneth Perkins
//    Date:   Aug 19, 2013
//    Taken From: http://programmingnotes.org/
//    File:  Parallel.cpp
//    Description: File downloaders are programs used for downloading files
//       from the Internet. Using the fork() & execlp("wget") command, the
//       following is a multi-process parallel file downloader which reads an
//       input file containing url file download links as a commandline
//       argument and downloads the files located on the internet all at once.
// ============================================================================
#include <iostream>
#include <cstring>
#include <fstream>
#include <cstdlib>
#include <unistd.h>
#include <sys/wait.h>
using namespace std;

// compile & run
// g++ Parallel.cpp -o Parallel
// ./Parallel urls.txt

int main(int argc, char* argv[])
{
    // declare variables
    pid_t pid = -1;
    int urlNumber = 0;
    char urlName[256];
    ifstream infile;

    // check if theres enough command line args
    if(argc < 2)
    {
        cout <<"\nERROR -- NOT ENOUGH ARGS!"
            <<"\n\nUSAGE: "<<argv[0]<<" <file containing url downloads>\n\n";
        exit(1);
    }

    // try to open the file containing the download url links
    // exit if the url file is not found
    infile.open(argv[1]);
    if(infile.fail())
    {
        cout <<"\nERROR -- "<<argv[1]<<" NOT FOUND!\n\n";
        exit(1);
    }

    // get download url links from the file
    while(infile.getline(urlName, sizeof(urlName)))
    {
        ++urlNumber;

        // fork another process
        pid = fork();

        if(pid < 0)
        {   // ** error occurred
            perror("fork");
            exit(1);
        }
        else if(pid == 0)
        {   // ** child process
            cout <<endl<<"** URL #"<<urlNumber
                    <<" is currently downloading... **\n\n";
            execlp("/usr/bin/wget", "wget", urlName, NULL);
        }
    }
    infile.close();

    while(urlNumber > 0)
    {   // ** parent process
        // parent will wait for the child to complete
        wait(NULL);
        cout <<endl<<"-- URL #"<<urlNumber<<" is complete! --\n";
        --urlNumber;
    }

    cout <<endl<<"The parent process is now exiting...\n";

    return 0;
}// http://programmingnotes.org/
			
				
					
				
					1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
				
						// ============================================================================
//    Author: Kenneth Perkins
//    Date:   Aug 19, 2013
//    Taken From: http://programmingnotes.org/
//    File:  Parallel.cpp
//    Description: File downloaders are programs used for downloading files
//       from the Internet. Using the fork() & execlp("wget") command, the
//       following is a multi-process parallel file downloader which reads an
//       input file containing url file download links as a commandline
//       argument and downloads the files located on the internet all at once.
// ============================================================================
#include <iostream>
#include <cstring>
#include <fstream>
#include <cstdlib>
#include <unistd.h>
#include <sys/wait.h>
using namespace std;
 
// compile & run
// g++ Parallel.cpp -o Parallel
// ./Parallel urls.txt
 
int main(int argc, char* argv[])
{
    // declare variables
    pid_t pid = -1;
    int urlNumber = 0;
    char urlName[256];
    ifstream infile;
 
    // check if theres enough command line args
    if(argc < 2)
    {
        cout <<"\nERROR -- NOT ENOUGH ARGS!"
            <<"\n\nUSAGE: "<<argv[0]<<" <file containing url downloads>\n\n";
        exit(1);
    }
 
    // try to open the file containing the download url links
    // exit if the url file is not found
    infile.open(argv[1]);
    if(infile.fail())
    {
        cout <<"\nERROR -- "<<argv[1]<<" NOT FOUND!\n\n";
        exit(1);
    }
 
    // get download url links from the file
    while(infile.getline(urlName, sizeof(urlName)))
    {
        ++urlNumber;
 
        // fork another process
        pid = fork();
 
        if(pid < 0)
        {   // ** error occurred
            perror("fork");
            exit(1);
        }
        else if(pid == 0)
        {   // ** child process
            cout <<endl<<"** URL #"<<urlNumber
                    <<" is currently downloading... **\n\n";
            execlp("/usr/bin/wget", "wget", urlName, NULL);
        }
    }
    infile.close();
 
    while(urlNumber > 0)
    {   // ** parent process
        // parent will wait for the child to complete
        wait(NULL);
        cout <<endl<<"-- URL #"<<urlNumber<<" is complete! --\n";
        --urlNumber;
    }
 
    cout <<endl<<"The parent process is now exiting...\n";
 
    return 0;
}// http://programmingnotes.org/

QUICK NOTES:
The highlighted lines are sections of interest to look out for.

The code is heavily commented, so no further insight is necessary. If you have any questions, feel free to leave a comment below.

Also note, while the parallel downloader executes, the outputs from different children may intermingle.

C++ || Snippet – How To Use Fork & Pipe For Interprocess Communication

August 19, 2013 admin No comments

The following is sample code which demonstrates the use of the fork, read, and write function calls for use with pipes on Unix based systems.

A pipe is a mechanism for interprocess communication. Data written to a pipe by one process can be read by another process. Creating a pipe is achieved by using the pipe function, which creates both the reading and writing ends of the pipe file descriptor.

In typical use, a parent process creates a pipe just before it forks one or more child processes. The pipe is then used for communication between either the parent or child processes, or between two sibling processes.

A real world example of this kind of communication can be seen in all operating system terminal shells. When you type a command in a shell, it will spawn the executable represented by that command with a call to fork. A pipe is opened to the new child process, and its output is read and printed by the terminal.





		
		
			Fork & Pipe Interprocess Communication
			
C++
			
			// ============================================================================
//    Author: Kenneth Perkins
//    Date:   Aug 19, 2013
//    Taken From: http://programmingnotes.org/
//    File:  ForkPipeExample.cpp
//    Description: Demonstrate the use of the fork() and pipe() command for
//        parent and child interprocess communication. 
// ============================================================================
#include <iostream>
#include <cstdlib>
#include <cstring>
#include <sys/wait.h>
#include <unistd.h>
using namespace std;

// global variables
const int BUFFER_SIZE = 256;
const int READ_END = 0;
const int WRITE_END = 1; 

int main() 
{
    // declare variables
    int fd[2]; // pipe file descriptor
    pid_t pid = -1;     
    char writeMsg[BUFFER_SIZE] = "Greetings From Your Parent!";
    char readMsg[BUFFER_SIZE];
    
    // create a pipe
    if(pipe(fd) < 0) 
    {
        perror("pipe");
        exit(1);
    }
    
    cout <<"\nParent is forking a child." << endl;

    // create a duplicate process of this current program
    pid = fork();
    
    // exit if something went wrong with the fork
    if(pid < 0) 
    {
        perror("fork");
        exit(1);
    }
    // this code only gets executed by the child process
    else if(pid == 0)
    {
        cout <<"\nStarting the child process..\n\n";
        // close the "write" end b/c we arent using it
        close(fd[WRITE_END]);
        // get data from the pipe using the read() function
        read(fd[READ_END], readMsg, sizeof(readMsg));
        // close the "read" end
        close(fd[READ_END]);
        cout <<"Message from the parent via the pipe: "<<readMsg<<endl;
    }
    // the parent process executes here 
    else 
    {
        cout <<"\nParent is now waiting for child id #"<<pid<<" to complete..\n";
        // close the "read" end b/c we arent using it
        close(fd[READ_END]);
        // write data on pipe
        write(fd[WRITE_END], writeMsg, strlen(writeMsg)+1);
        // close the "write" end
        close(fd[WRITE_END]); 
        // use the "wait" function to wait for the child to complete
        wait(NULL);   
        cout <<"\nThe child process is complete and has terminated!\n";
    }
    
    // NOTE: this message gets displayed twice - Why?!
    cout <<"\nProgram is now exiting...\n";
    
    return 0;
}// http://programmingnotes.org/
			
				
					
				
					1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
				
						// ============================================================================
//    Author: Kenneth Perkins
//    Date:   Aug 19, 2013
//    Taken From: http://programmingnotes.org/
//    File:  ForkPipeExample.cpp
//    Description: Demonstrate the use of the fork() and pipe() command for
//        parent and child interprocess communication. 
// ============================================================================
#include <iostream>
#include <cstdlib>
#include <cstring>
#include <sys/wait.h>
#include <unistd.h>
using namespace std;
 
// global variables
const int BUFFER_SIZE = 256;
const int READ_END = 0;
const int WRITE_END = 1; 
 
int main() 
{
    // declare variables
    int fd[2]; // pipe file descriptor
    pid_t pid = -1;     
    char writeMsg[BUFFER_SIZE] = "Greetings From Your Parent!";
    char readMsg[BUFFER_SIZE];
    
    // create a pipe
    if(pipe(fd) < 0) 
    {
        perror("pipe");
        exit(1);
    }
    
    cout <<"\nParent is forking a child." << endl;
 
    // create a duplicate process of this current program
    pid = fork();
    
    // exit if something went wrong with the fork
    if(pid < 0) 
    {
        perror("fork");
        exit(1);
    }
    // this code only gets executed by the child process
    else if(pid == 0)
    {
        cout <<"\nStarting the child process..\n\n";
        // close the "write" end b/c we arent using it
        close(fd[WRITE_END]);
        // get data from the pipe using the read() function
        read(fd[READ_END], readMsg, sizeof(readMsg));
        // close the "read" end
        close(fd[READ_END]);
        cout <<"Message from the parent via the pipe: "<<readMsg<<endl;
    }
    // the parent process executes here 
    else 
    {
        cout <<"\nParent is now waiting for child id #"<<pid<<" to complete..\n";
        // close the "read" end b/c we arent using it
        close(fd[READ_END]);
        // write data on pipe
        write(fd[WRITE_END], writeMsg, strlen(writeMsg)+1);
        // close the "write" end
        close(fd[WRITE_END]); 
        // use the "wait" function to wait for the child to complete
        wait(NULL);   
        cout <<"\nThe child process is complete and has terminated!\n";
    }
    
    // NOTE: this message gets displayed twice - Why?!
    cout <<"\nProgram is now exiting...\n";
    
    return 0;
}// http://programmingnotes.org/

QUICK NOTES:
The highlighted lines are sections of interest to look out for.

The code is heavily commented, so no further insight is necessary. If you have any questions, feel free to leave a comment below.

The following is sample output:

Parent is forking a child. Parent is now waiting for child id #12776 to complete..
Starting the child process.. Message from the parent via the pipe: Greetings From Your Parent! Program is now exiting... The child process is complete and has terminated!
Program is now exiting...

C++ || Snippet – How To Use Fork & Execlp For Interprocess Communication

August 19, 2013 admin No comments

The following is sample code which demonstrates the use of the “fork” and “execlp” function calls on Unix based systems.

The “fork” function call creates a new process by duplicating the calling process; or in more simpler terms, it creates a duplicate process (a child) of the calling (parent) process.

This new process, referred to as the child, is an exact duplicate of the calling process, referred to as the parent.

The “execlp” function call is part of a family of functions which replaces a current running process image with a new process image. That means by using the “execlp” function call, you are basically replacing the entire current running process with a new user defined program.

Though fork and execlp are not required to be used together, they are often used in conjunction with one another as a way of creating a new program running as a child of another process.





		
		
			How To Use Fork & Execlp
			
C++
			
			// ============================================================================
//    Author: Kenneth Perkins
//    Date:   Aug 19, 2013
//    Taken From: http://programmingnotes.org/
//    File:  ForkExample.cpp
//    Description: Demonstrate the use of the fork() command for parent and
//       child process manipulation.
// ============================================================================
#include <iostream>
#include <cstdlib>
#include <sys/wait.h>
#include <unistd.h>
using namespace std;

int main()
{
    // declare variable
    pid_t pid = -1;

    cout <<"\nParent is forking a child." << endl;

    // create a duplicate process of this current program
    pid = fork();

    // exit if something went wrong with the fork
    if(pid < 0)
    {
        perror("fork");
        exit(1);
    }
    // this code only gets executed by the child process
    else if(pid == 0)
    {
        cout <<"\nStarting the child process..\n" << endl;
        // NOTE: the "execlp" function replaces the current
        // child process with the terminal "ls" command.
        // the child process finally dies after the "execlp"
        // function is complete
        execlp("/bin/ls", "ls", "-l", NULL); // "ls -l" command
        cout <<"\nNOTE: This never gets executed - Why?!";
    }
    // the parent process executes here
    else
    {
        cout <<"\nParent is now waiting for child id #"<<pid<<" to complete.." << endl;
        // use the "wait" function to wait for the child to complete
        wait(NULL);
        cout <<"\nThe child process is complete and has terminated!\n";
    }

    cout <<"\nParent is now exiting...\n";

    return 0;
}// http://programmingnotes.org/
			
				
					
				
					1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
				
						// ============================================================================
//    Author: Kenneth Perkins
//    Date:   Aug 19, 2013
//    Taken From: http://programmingnotes.org/
//    File:  ForkExample.cpp
//    Description: Demonstrate the use of the fork() command for parent and
//       child process manipulation.
// ============================================================================
#include <iostream>
#include <cstdlib>
#include <sys/wait.h>
#include <unistd.h>
using namespace std;
 
int main()
{
    // declare variable
    pid_t pid = -1;
 
    cout <<"\nParent is forking a child." << endl;
 
    // create a duplicate process of this current program
    pid = fork();
 
    // exit if something went wrong with the fork
    if(pid < 0)
    {
        perror("fork");
        exit(1);
    }
    // this code only gets executed by the child process
    else if(pid == 0)
    {
        cout <<"\nStarting the child process..\n" << endl;
        // NOTE: the "execlp" function replaces the current
        // child process with the terminal "ls" command.
        // the child process finally dies after the "execlp"
        // function is complete
        execlp("/bin/ls", "ls", "-l", NULL); // "ls -l" command
        cout <<"\nNOTE: This never gets executed - Why?!";
    }
    // the parent process executes here
    else
    {
        cout <<"\nParent is now waiting for child id #"<<pid<<" to complete.." << endl;
        // use the "wait" function to wait for the child to complete
        wait(NULL);
        cout <<"\nThe child process is complete and has terminated!\n";
    }
 
    cout <<"\nParent is now exiting...\n";
 
    return 0;
}// http://programmingnotes.org/

QUICK NOTES:
The highlighted lines are sections of interest to look out for.

The code is heavily commented, so no further insight is necessary. If you have any questions, feel free to leave a comment below.

The following is sample output:

Parent is forking a child. Parent is now waiting for child id #10872 to complete..
Starting the child process.. total 1466 -rw-r--r-- 1 admin admin 468 Apr 27 2012 nautilus-computer.desktop -rwxrwxr-x 1 admin admin 9190 Aug 19 15:17 ForkExample -rw-rw-r-- 1 admin admin 1640 Aug 19 15:17 ForkExample.cpp The child process is complete and has terminated!
Parent is now exiting...

Tag Archives: fork

C++ || Snippet – How To Use Message Queues For Interprocess Communication

C++ || Snippet – How To Create And Use Threads For Interprocess Communication

C++ || Multi-Hash Interprocess Communication Using Fork, Popen, & Pipes

C++ || Serial & Parallel Multi Process File Downloader Using Fork & Execlp

C++ || Snippet – How To Use Fork & Pipe For Interprocess Communication

C++ || Snippet – How To Use Fork & Execlp For Interprocess Communication

Search

Compile Your Code Now

Translate Page

Popular Pages

Recent

Archives

Categories