Deadlock detection algorithm source code c

CSci 430: Programming Project #3 Deadlock Detection

prog-03.pdf
CSci 430: Programming Project #3

Deadlock Detection

Spring 2019

Dates:

Assigned: Monday February 25, 2019 Due: Wednesday March 13, 2019 (before Midnight)

Objectives:

ˆ Learn more about Deadlock algorithms.

ˆ Better understand how we can algorithmically detect deadlocks on a system.

ˆ Use C/C++ to implement vector and matrix data structures, get prac- tice in creating and using such data structures in C/C++.

Description:

Our textbook gives the following algorithm (pg. 276) for algorithmically detecting if a deadlock is present or not in a system. It requires that the system keep an Allocation matrix A, listing which resources are currently allocated to which processes, and the available vector V, which gives the amount of each resource currently available in the system. In addition, the deadlock detection algorithm requies a request matrix Q, which keeps track of the amount of each resource each process is currently requesting from the system. The algorithm is:

1. Mark each process that has a row in the Allocation matrix of all zeros.

2. Initialize a temporary vector W to equal the Available vector A.

1

3. Find an index i such that process i is currently unmarked and the i th row of Q is less than or equal toW. That is, Qik ≤ Wk, for 1 ≤ k ≤ m. If no such row is found, terminate the algorithm.

4. If such a row is found, mark process i and add the corresponding row of the allocation matrix toW. That is, setWk = Wk+Aik, for 1 ≤ k ≤ m. Return to step 3.

A deadlock exists if and only if there are unmarked processes at the end of the algorithm. Each unmarked process is deadlocked.

In this assignment we will implement the deadlock detection algorithm. Your program will be given a �le that describes the A allocation matrix and the Q request matrix, representing the current state of all allocations and requested allocations in the system. Your program will implement the deadlock detection algorithm described above. The result of your program will be one of 2 outputs:

1. If no deadlock exists, the program will display No Deadlock on stan- dard output.

2. If a deadlock does exist, the program will display Deadlock: P0, P1, P2 on standard output, where P0, P1, P2 are the processes that the algorithm determined to be deadlocked in the system.

State simulation �le formats

I have provided a p3-start.cpp template that can open up and read in the process/resource state simulation �les used for this assignment. Here we discuss a bit more the format of these �le. I have provided 2 or 3 exam- ple simulations, with expected correct answers, for you to use to test your implementations with.

The input �les needed for this assignment need to contain the information found in theV available vector and theA allocation andQ request matrices. In the following I use r as the number of resources and p as the number of processes. Thus the general format of the input �le is:

r p

V1 V2 V3 ... Vr

A11 A12 ... A1r

...

Ap1 Ap2 ... Apr

2

Q11 Q12 ... Q1r

...

Qp1 Qp2 ... Qpr

For example, the example of the deadlock detection algorithm given on page 277 has a system with r=5 resources and p=4 processes. The V, A and Q vector/matrices are shown on that page. The input �le for the current state of the system shown on page 277 would be

5 4

0 0 0 0 1

1 0 1 1 0

1 1 0 0 0

0 0 0 1 0

0 0 0 0 0

0 1 0 0 1

0 0 1 0 1

0 0 0 0 1

1 0 1 0 1

The function named readSystemState() in your template p2-start.cpp code expects a �le of this format, and reads it into a State structure for you.

Running Simulations

The following is a discussion of the expected output of your program. Your program must work from the command line, and expect a single parameter, the name of the state simulation input �le, as its input. Your program should display only a single line to standard output as a result of running it. If the system, described in the state input �le is not deadlocked, the program should simply state there was no deadlock to standard output:

$ p3.exe state-02.sim

No Deadlock

On the other hand, if your program is deadlocked, it should say that it detected a deadlock, and it should print out the processes that are deadloked to standard output:

$ p3.exe state-01.sim

Deadlock: P0, P1,

3

I have provided 2 or 3 example input state �les, named state-01.sim, state-02.sim, etc. I have also provided the correct and expected output for these simulations, named state-01.res, state-02.out, etc.

4

prog-03.zip
p3-start.cpp
p3-start.cpp
/** @file p3-start.cpp
*
* @author Your Name Here
*
* @assg Programming Assignment #3
*
* @desc Implement the deadlock detection algorithm. Given a file
* that describes the current allocation A of resources in the
* system, and the current set of outstanding requests Q in
* the system, determine if a deadlock is present or not. Use
* the algorithm given on p.276 in the Stallings textbook.
*
* @date March 05, 2018
*/
#include < stdlib . h >
#include < iostream >
#include < iomanip >
#include < fstream >
#include < string >
using namespace std ;

// global constants
const int MAX_PROCESSES = 10 ; // I won't test your algorithm with simulations with more than 10 processes
const int MAX_RESOURCES = 10 ; // nor will I give a simulation to test with more than 10 resources

// simple struct to read in and hold the state of the system
typedef struct
{
int numResources ;
int numProcesses ;
int available [ MAX_RESOURCES ]; // V available vector
int alloc [ MAX_PROCESSES ][ MAX_RESOURCES ]; // A allocation matrix
int request [ MAX_PROCESSES ][ MAX_RESOURCES ]; // Q request matrix
} State ;

/** Read system state from file.
* Given a file, read the current system state from the file.
* The system state file is expected to hold the available vector V
* the allocation matrix A and the request matrix Q.
*
* @param simfilename The name of the file to open and read state & request
* from.
* @return state A new State structure is allocated and filled with the
* system state from the file. A pointer to this allocated system
* state structure is returned as a result of calling this function.
*/
State * readSystemState ( char * statefilename )
{
ifstream simstatefile ( statefilename );
State * state ;
int r , p ;

// If we can't open file, abort and let the user know problem
if ( ! simstatefile . is_open ())
{
cout << "Error: could not open system state file: " << statefilename
<< endl ;
exit ( 1 );
}

// dynamically allocate a new State structure, to be filled in and returned
state = new State ;

// Format of file is this (where m = numResource n = numProcesses
// V = available vector
// A = allocation matrix and
// Q = request matrix)
// m n
// V1 V2 V3 ... Vm
// A11 A12 ... A1m
// ...
// An1 An2 ... Anm
// Q11 Q12 ... Q1m
// ...
// Qn1 Qn2 ... Qnm

// First line, get m (numResources) and n (numProcesses)
simstatefile >> state -> numResources >> state -> numProcesses ;

// Next line contains the available vector V
for ( r = 0 ; r < state -> numResources ; r ++ )
{
simstatefile >> state -> available [ r ];
}

// Next n lines contain the allocation matrix A
for ( p = 0 ; p < state -> numProcesses ; p ++ )
{
for ( r = 0 ; r < state -> numResources ; r ++ )
{
simstatefile >> state -> alloc [ p ][ r ];
}
}

// Next n lines contain the request matrix Q
for ( p = 0 ; p < state -> numProcesses ; p ++ )
{
for ( r = 0 ; r < state -> numResources ; r ++ )
{
simstatefile >> state -> request [ p ][ r ];
}
}

// return the newly allocated and filled in system state
return state ;
}

/** Display a vector
* Display a state vector to standard output
*
* @param len The number of items in the vector
* @param v An array of integers of len items
*/
void displayVector ( int len , int v [])
{
int i ;

// Display a header
for ( i = 0 ; i < len ; i ++ )
{
cout << "R" << i << " " ;
}
cout << endl ;

// Display values
for ( i = 0 ; i < len ; i ++ )
{
cout << setw ( 2 ) << v [ i ] << " " ;
}
cout << endl ;
}

/** Display a matrix
* Display a state matrix to standard output
*
* @param rows The number of rows in the matrix
* @param cols The number of cols in the matrix
* @param m A 2 dimensional array of rows x cols integers
*/
void displayMatrix ( int rows , int cols , int v [ MAX_PROCESSES ][ MAX_RESOURCES ])
{
int r , c ;

// display column headers
cout << " " ; // extra space over for row labels
for ( c = 0 ; c < cols ; c ++ )
{
cout << "R" << c << " " ;
}
cout << endl ;

// now display data in matrix
for ( r = 0 ; r < rows ; r ++ )
{
cout << "P" << r << " " ;
for ( c = 0 ; c < cols ; c ++ )
{
cout << setw ( 2 ) << v [ r ][ c ] << " " ;
}
cout << endl ;
}
cout << endl ;
}

/** Display state
* Display the values of the resource vectors and matrices in the indicated
* state structure
*
* @param state A pointer to a State struct whose info we should display on stdout.
*/
void displayState ( State * s )
{
cout << "numResources (m) = " << s -> numResources << " " ;
cout << "numProcesses (n) = " << s -> numProcesses << endl << endl ;

cout << "Available vector V:" << endl ;
displayVector ( s -> numResources , s -> available );
cout << endl ;

cout << "Allocation matrix A: " << endl ;
displayMatrix ( s -> numProcesses , s -> numResources , s -> alloc );
cout << endl ;

cout << "Request matrix Q: " << endl ;
displayMatrix ( s -> numProcesses , s -> numResources , s -> request );
cout << endl ;

}

/** The deadlock detector
* The starting point for implementation of the deadlock detection algorithm.
* We open and read in the allocation matrices here, then perform the deadlock detection.
*
* @ param statefilename A string with the name of the file holding the A and Q system state matrices
*/
void detectDeadlock ( char * statefilename )
{
State * state ;

state = readSystemState ( statefilename );
// I have provided some example routines to read and display system state, implemented as a plain
// C struct using C 1 and 2 dimensional arrays. You can uncomment out the following, and/or use
// the displayMatrix() and displayVector() functions to help you debug. But make sure you
// remove or comment back up any statements after you are done debugging.
displayState ( state );

// You need to implement your solution here. I would recommend you use functions for each of
// these steps.

// Step 1: Set up a data structure that records marked/unmarked
// processes. All processes are initially unmarked Search
// through the allocation matrix to find rows of all 0, and
// mark corresponding processes in your mark structure

// Step 2: Create a temporary vector W. Copy contents of available
// vector V to W. Suggestion: create a function called
// copyVector, that takes a vector as its parameter, and returns
// a new vector.

// Need to put Steps 3 and 4 in a loop

// Step 3: Find index i such that process i is currently unmarked,
// and the ith row of Q is less than or equal to W. If no
// such process is found, need to terminate algorithm/loop.
// Suggestion: write a function that takes Q and W, and
// returns either i (index of process meeting criteria) or
// -1

// Step 4: If a row was found (e.g. i was a valid process that met
// criteria of step 3), mark process i and add the
// correspoinding row of allocation matrix to W. Loop back
// to beginning of step 3.

// Step 5: after loop finishes,
// if (your marked/unmarked processes contains unmarked processes)
// {
// cout << "Deadlock";
// // loop through your marked/unmarked structure, print out all unmarked processes as P1, P2, etc.
// cout << endl;
// }
// else // all processes were marked, so no deadlock
// {
// cout << "No Deadlock" << endl;
// }
}

/** Main entry point of deadlock detection.
* The main entry point of the deadlock detection program. This function
* checks the command line arguments, and calls the detection function if correct
* arguments were supplied. We expect a single command line argument
* which is the name of the file holding the allocation and request matrices
* of the current state of the system.
*
* @param argc The argument count
* @param argv The command line argument values. We expect argv[1] to be the
* name of a file in the current directory holding A and Q matrices.
*/
int main ( int argc , char ** argv )
{
if ( argc != 2 )
{
cout << "Error: expecting state matrix file as first command line parameter" << endl ;
cout << "Usage: " << argv [ 0 ] << " system-state.sim" << endl ;
exit ( 1 );
}

detectDeadlock ( argv [ 1 ]);
// if don't want to use command line do following. Need to recompile by hand since file
// name to get simulated events from is hard coded.
// Make sure you revert back to using command line before submitting your program.
//detectDeadlock("state-01.sim");
}
prog-03.pdf
CSci 430: Programming Project #3

Deadlock Detection

Spring 2019

Dates:

Assigned: Monday February 25, 2019 Due: Wednesday March 13, 2019 (before Midnight)

Objectives:

ˆ Learn more about Deadlock algorithms.

ˆ Better understand how we can algorithmically detect deadlocks on a system.

ˆ Use C/C++ to implement vector and matrix data structures, get prac- tice in creating and using such data structures in C/C++.

Description:

Our textbook gives the following algorithm (pg. 276) for algorithmically detecting if a deadlock is present or not in a system. It requires that the system keep an Allocation matrix A, listing which resources are currently allocated to which processes, and the available vector V, which gives the amount of each resource currently available in the system. In addition, the deadlock detection algorithm requies a request matrix Q, which keeps track of the amount of each resource each process is currently requesting from the system. The algorithm is:

1. Mark each process that has a row in the Allocation matrix of all zeros.

2. Initialize a temporary vector W to equal the Available vector A.

1

3. Find an index i such that process i is currently unmarked and the i th row of Q is less than or equal toW. That is, Qik ≤ Wk, for 1 ≤ k ≤ m. If no such row is found, terminate the algorithm.

4. If such a row is found, mark process i and add the corresponding row of the allocation matrix toW. That is, setWk = Wk+Aik, for 1 ≤ k ≤ m. Return to step 3.

A deadlock exists if and only if there are unmarked processes at the end of the algorithm. Each unmarked process is deadlocked.

In this assignment we will implement the deadlock detection algorithm. Your program will be given a �le that describes the A allocation matrix and the Q request matrix, representing the current state of all allocations and requested allocations in the system. Your program will implement the deadlock detection algorithm described above. The result of your program will be one of 2 outputs:

1. If no deadlock exists, the program will display No Deadlock on stan- dard output.

2. If a deadlock does exist, the program will display Deadlock: P0, P1, P2 on standard output, where P0, P1, P2 are the processes that the algorithm determined to be deadlocked in the system.

State simulation �le formats

I have provided a p3-start.cpp template that can open up and read in the process/resource state simulation �les used for this assignment. Here we discuss a bit more the format of these �le. I have provided 2 or 3 exam- ple simulations, with expected correct answers, for you to use to test your implementations with.

The input �les needed for this assignment need to contain the information found in theV available vector and theA allocation andQ request matrices. In the following I use r as the number of resources and p as the number of processes. Thus the general format of the input �le is:

r p

V1 V2 V3 ... Vr

A11 A12 ... A1r

...

Ap1 Ap2 ... Apr

2

Q11 Q12 ... Q1r

...

Qp1 Qp2 ... Qpr

For example, the example of the deadlock detection algorithm given on page 277 has a system with r=5 resources and p=4 processes. The V, A and Q vector/matrices are shown on that page. The input �le for the current state of the system shown on page 277 would be

5 4

0 0 0 0 1

1 0 1 1 0

1 1 0 0 0

0 0 0 1 0

0 0 0 0 0

0 1 0 0 1

0 0 1 0 1

0 0 0 0 1

1 0 1 0 1

The function named readSystemState() in your template p2-start.cpp code expects a �le of this format, and reads it into a State structure for you.

Running Simulations

The following is a discussion of the expected output of your program. Your program must work from the command line, and expect a single parameter, the name of the state simulation input �le, as its input. Your program should display only a single line to standard output as a result of running it. If the system, described in the state input �le is not deadlocked, the program should simply state there was no deadlock to standard output:

$ p3.exe state-02.sim

No Deadlock

On the other hand, if your program is deadlocked, it should say that it detected a deadlock, and it should print out the processes that are deadloked to standard output:

$ p3.exe state-01.sim

Deadlock: P0, P1,

3

I have provided 2 or 3 example input state �les, named state-01.sim, state-02.sim, etc. I have also provided the correct and expected output for these simulations, named state-01.res, state-02.out, etc.

4

state-01.sim
5 4 0 0 0 0 1 1 0 1 1 0 1 1 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 1 0 0 1 0 1 0 0 0 0 1 1 0 1 0 1