MODULE 3 :PROCESS SYNCHRONIZATION,DEADLOCKS

Process Synchronization
2.1 Synchronization
2.2 The Critical-Section Problem
2.3 Peterson‘s Solution
2.4 Synchronization Hardware
2.4.1 Hardware based Solution for Critical-section Problem
2.4.2 Hardware instructions for solving critical-section problem
2.4.2.1 TestAndSet()
2.4.2.2 TestAndSet with Mutual Exclusion
2.4.2.3 Swap()
2.4.2.4 Bounded waiting Mutual Exclusion with TestAndSet()
2.5 Semaphores
2.5.1 Semaphore Usage
2.5.2 Semaphore Implementation
2.5.3 Deadlocks & Starvation
2.6 Classic Problems of Synchronization
2.6.1 Bounded-Buffer Problem
2.6.2 Readers-Writers Problem
2.6.3 Dining-Philosophers Problem

3.1 Deadlocks
3.2 System Model
3.3 Deadlock Characterization
3.3.1 Necessary Conditions
3.3.2 Resource Allocation Graph
3.4 Methods for Handling Deadlocks
3.5 Deadlock Prevention
3.5.1 Mutual Exclusion
3.5.2 Hold and Wait
3.5.3 No Preemption
3.5.4 Circular Wait
3.6 Deadlock Avoidance
3.6.1 Safe State
3.6.2 Resource Allocation Graph Algorithm
3.6.3 Banker's Algorithm
3.6.3.1 Safety Algorithm
3.6.3.2 Resource Request Algorithm
3.6 3.3 An Illustrative Example
3.7 Deadlock Detection
3.7.1 Single Instance of Each Resource Type
3.7.2 Several Instances of a Resource Type
3.7.3 Detection Algorithm Usage
3.8 Recovery from Deadlock
3.8.1 Process Termination
3.8.2 Resource Preemption

2.1 Synchronization
• Co-operating process is one that can affect or be affected by other processes.
• Co-operating processes may either
→ share a logical address-space (i.e. code & data) or
→ share data through files or
→ messages through threads.
• Concurrent-access to shared-data may result in data-inconsistency.
• To maintain data-consistency:
The orderly execution of co-operating processes is necessary.
• Suppose that we wanted to provide a solution to producer-consumer problem that fills all buffers.
We can do so by having an variable counter that keeps track of the no. of full buffers.
Initially, counter=0.
 counter is incremented by the producer after it produces a new buffer.
 counter is decremented by the consumer after it consumes a buffer.
• Shared-data:

Producer Process: Consumer Process:

• A situation where several processes access & manipulate same data concurrently and the outcome of
the execution depends on particular order in which the access takes place, is called a race condition.
• Example:
counter++ could be implemented as: counter— may be implemented as:

• Consider this execution interleaving with counter = 5 initially:

• The value of counter may be either 4 or 6, where the correct result should be 5. This is an example
for race condition.
• To prevent race conditions, concurrent-processes must be synchronized.
 2.2 Critical-Section Problem
• Critical-section is a segment-of-code in which a process may be
→ changing common variables
→ updating a table or
→ writing a file.
• Each process has a critical-section in which the shared-data is accessed.
• General structure of a typical process has following (Figure 2.12):
1) Entry-section
 Requests permission to enter the critical-section.
2) Critical-section
 Mutually exclusive in time i.e. no other process can execute in its critical-section.
3) Exit-section
 Follows the critical-section.
4) Remainder-section

Figure 2.12 General structure of a typical process

• Problem statement:
―Ensure that when one process is executing in its critical-section, no other process is to be allowed
to execute in its critical-section‖.
• A solution to the problem must satisfy the following 3 requirements:
1) Mutual Exclusion
 Only one process can be in its critical-section.
2) Progress
 Only processes that are not in their remainder-section can enter their critical-section, and the
selection of a process cannot be postponed indefinitely.
3) Bounded Waiting
 There must be a bound on the number of times that other processes are allowed to enter
their critical-sections after a process has made a request to enter its critical-section and before
the request is granted.
• Two approaches used to handle critical-sections:
1) Preemptive Kernels
 Allows a process to be preempted while it is running in kernel-mode.
2) Non-preemptive Kernels
 Does not allow a process running in kernel-mode to be preempted.
 2.3 Peterson’s Solution
• This is a classic software-based solution to the critical-section problem.
• This is limited to 2 processes.
• The 2 processes alternate execution between
→ critical-sections and
→ remainder-sections.
• The 2 processes share 2 variables (Figure 2.13):

where turn = indicates whose turn it is to enter its critical-section.

(i.e., if turn==i, then process Pi is allowed to execute in its critical-section).
flag = used to indicate if a process is ready to enter its critical-section.
(i.e. if flag[i]=true, then Pi is ready to enter its critical-section).

Figure 2.13 The structure of process Pi in Peterson‘s solution

• To enter the critical-section,

→ firstly process Pi sets flag[i] to be true and
→ then sets turn to the value j.
• If both processes try to enter at the same time, turn will be set to both i and j at roughly the same
time.
• The final value of turn determines which of the 2 processes is allowed to enter its critical-section first.
• To prove that this solution is correct, we show that:
1) Mutual-exclusion is preserved.
2) The progress requirement is satisfied.
3) The bounded-waiting requirement is met.
 2.4 Synchronization Hardware
2.4.1 Hardware based Solution for Critical-section Problem
• A lock is a simple tool used to solve the critical-section problem.
• Race conditions are prevented by following restriction (Figure 2.14).
―A process must acquire a lock before entering a critical-
section. The process releases the lock when it exits the critical-

section‖.
Figure 2.14: Solution to the critical-section problem using locks

2.4.2 Hardware instructions for solving critical-section problem

• Modern systems provide special hardware instructions
→ to test & modify the content of a word atomically or
→ to swap the contents of 2 words atomically.
• Atomic-operation means an operation that completes in its entirety without interruption.

2.13.2.1 TestAndSet()
• This instruction is executed atomically (Figure 2.15).
• If two TestAndSet() are executed simultaneously (on different CPU), they will be executed
sequentially in some arbitrary order.

Figure 2.15 The definition of the test and set() instruction

2.13.2.2 TestAndSet with Mutual Exclusion

• If the machine supports the TestAndSet(), we can implement mutual-exclusion by declaring a
boolean variable lock, initialized to false (Figure 2.16).

Figure 2.16 Mutual-exclusion implementation with test and set()

2.13.2.3 Swap()
• This instruction is executed atomically (Figure 2.17).
• If the machine supports the Swap(), then mutual-exclusion can be provided as follows:
1) A global boolean variable lock is declared and is initialized to false.
2) In addition, each process has a local Boolean variable key (Figure 2.18).

Figure 2.17 The definition of swap() instruction

Figure 2.18 Mutual-exclusion implementation with the swap() instruction

2.13.2.4 Bounded waiting Mutual Exclusion with TestAndSet()

• Common data structures are

• These data structures are initialized to false (Figure 2.19).

Figure 2.19 Bounded-waiting mutual-exclusion with TestandSet()

2.5 Semaphores
• A semaphore is a synchronization-tool.
• It used to control access to shared-variables so that only one process may at any point in time
change the value of the shared-variable.
• A semaphore(S) is an integer-variable that is accessed only through 2 atomic-operations:
1) wait() and
2) signal().
• wait() is termed P ("to test").
signal() is termed V ("to increment").
• Definition of wait(): Definition of signal():

• When one process modifies the semaphore-value, no other process can simultaneously modify that
same semaphore-value.
• Also, in the case of wait(S), following 2 operations must be executed without interruption:
1) Testing of S(S<=0) and
2) Modification of S (S--)

2.14.1 Semaphore
Usage Counting
Semaphore
• The value of a semaphore can range over an unrestricted domain
Binary Semaphore
• The value of a semaphore can range only between 0 and 1.
• On some systems, binary semaphores are known as mutex locks, as they are locks that provide
mutual-exclusion.

1) Solution for Critical-section Problem using Binary Semaphores

• Binary semaphores can be used to solve the critical-section problem for multiple processes.
• The ‗n‘ processes share a semaphore mutex initialized to 1 (Figure 2.20).

Figure 2.20 Mutual-exclusion implementation with semaphores

2) Use of Counting Semaphores

• Counting semaphores can be used to control access to a given resource consisting of a finite number
o£ instances.
• The semaphore is initialized to the number of resources available.
• Each process that wishes to use a resource performs a wait() operation on the semaphore (thereby
decrementing the count).
• When a process releases a resource, it performs a signal() operation (incrementing the count).
• When the count for the semaphore goes to 0, all resources are being used.
• After that, processes that wish to use a resource will block until the count becomes greater than 0.

3) Solving Synchronization Problems

• Semaphores can also be used to solve synchronization problems.
• For example, consider 2 concurrently running-processes:
P1 with a statement S1 and
P2 with a statement S2.
• Suppose we require that S2 be executed only after S1 has completed.
• We can implement this scheme readily
→ by letting P1 and P2 share a common semaphore synch initialized to 0, and
→ by inserting the following statements in process P1

and the following statements in process P2

• Because synch is initialized to 0, P2 will execute S2 only after P1 has invoked signal (synch), which is
after statement S1 has been executed.

2.14.2 Semaphore Implementation

• Main disadvantage of semaphore: Busy waiting.
• Busy waiting: While a process is in its critical-section, any other process that tries to enter its
critical-section must loop continuously in the entry code.
• Busy waiting wastes CPU cycles that some other process might be able to use productively.
• This type of semaphore is also called a spinlock (because the process "spins" while waiting for the
lock).
• To overcome busy waiting, we can modify the definition of the wait() and signal() as follows:
1) When a process executes the wait() and finds that the semaphore-value is not positive, it
must wait. However, rather than engaging in busy waiting, the process can block itself.
2) A process that is blocked (waiting on a semaphore S) should be restarted when some other
process executes a signal(). The process is restarted by a wakeup().
• We assume 2 simple operations:
1) block() suspends the process that invokes it.
2) wakeup(P) resumes the execution of a blocked process P.
• We define a semaphore as follows:

• Definition of wait(): Definition of signal():

• This (critical-section) problem can be solved in two ways:

1) In a uni-processor environment
¤ Inhibit interrupts when the wait and signal operations execute.
¤ Only current process executes, until interrupts are re-enabled & the scheduler regains
control.
2) In a multiprocessor environment
¤ Inhibiting interrupts doesn't work.
¤ Use the hardware / software solutions described above.

2.14.3 Deadlocks & Starvation

• Deadlock occurs when 2 or more processes are waiting indefinitely for an event that can be caused
by only one of the waiting processes.
• The event in question is the execution of a signal() operation.
• To illustrate this, consider 2 processes, Po and P1, each accessing 2 semaphores, S and Q. Let S and
Q be initialized to 1.
• Suppose that Po executes wait(S) and then P1 executes wait(Q).
When Po executes wait(Q), it must wait until P1 executes signal(Q).
Similarly, when P1 executes wait(S), it must wait until Po executes signal(S).
Since these signal() operations cannot be executed, Po & P1 are deadlocked.
• Starvation (indefinite blocking) is another problem related to deadlocks.
• Starvation is a situation in which processes wait indefinitely within the semaphore.
• Indefinite blocking may occur if we remove processes from the list associated with a semaphore in
LIFO (last-in, first-out) order
 2.6_Classic Problems of Synchronization
1) Bounded-Buffer Problem
2) Readers and Writers Problem
3) Dining-Philosophers Problem

2.6.1 The Bounded-Buffer Problem

• The bounded-buffer problem is related to the producer consumer problem.
• There is a pool of n buffers, each capable of holding one item.
• Shared-data

where,
¤ mutex provides mutual-exclusion for accesses to the buffer-pool.
¤ empty counts the number of empty buffers.
¤ full counts the number of full buffers.
• The symmetry between the producer and the consumer.
¤ The producer produces full buffers for the consumer.
¤ The consumer produces empty buffers for the producer.
• Producer Process: Consumer Process:
 2.6.2 The Readers-Writers Problem
• A data set is shared among a number of concurrent processes.
• Readers are processes which want to only read the database (DB).
Writers are processes which want to update (i.e. to read & write) the DB.
• Problem:
 Obviously, if 2 readers can access the shared-DB simultaneously without any problems.
 However, if a writer & other process (either a reader or a writer) access the shared-DB
simultaneously, problems may arise.
Solution:
 The writers must have exclusive access to the shared-DB while writing to the DB.
• Shared-data

where,
¤ mutex is used to ensure mutual-exclusion when the variable readcount is updated.
¤ wrt is common to both reader and writer processes.
wrt is used as a mutual-exclusion semaphore for the writers.
wrt is also used by the first/last reader that enters/exits the critical-section.
¤ readcount counts no. of processes currently reading the object.
Initialization
mutex = 1, wrt = 1, readcount = 0
Writer Process: Reader Process:

• The readers-writers problem and its solutions are used to provide reader-writer locks on some
systems.
• The mode of lock needs to be specified:
2.6.2.1 read mode
 When a process wishes to read shared-data, it requests the lock in read mode.
2.6.2.2 write mode
 When a process wishes to modify shared-data, it requests the lock in write mode.
• Multiple processes are permitted to concurrently acquire a lock in read mode,
but only one process may acquire the lock for writing.
• These locks are most useful in the following situations:
1) In applications where it is easy to identify
→ which processes only read shared-data and
→ which threads only write shared-data.
2) In applications that have more readers than writers.
 2.6.3 The Dining-Philosophers Problem
• Problem statement:
 There are 5 philosophers with 5 chopsticks (semaphores).
 A philosopher is either eating (with two chopsticks) or thinking.
 The philosophers share a circular table (Figure 2.21).
 The table has
→ a bowl of rice in the center and
→ 5 single chopsticks.
 From time to time, a philosopher gets hungry and tries to pick up the 2 chopsticks that are
closest to her.
 A philosopher may pick up only one chopstick at a time.
 Obviously, she cannot pick up a chopstick that is already in the hand of a neighbor.
 When hungry philosopher has both her chopsticks at the same time, she eats without
releasing her chopsticks.
 When she is finished eating, she puts down both of her chopsticks and starts thinking again.
• Problem objective:
To allocate several resources among several processes in a deadlock-free & starvation-free manner.
• Solution:
 Represent each chopstick with a semaphore (Figure 2.22).
 A philosopher tries to grab a chopstick by executing a wait() on the semaphore.
 The philosopher releases her chopsticks by executing the signal() on the semaphores.
 This solution guarantees that no two neighbors are eating simultaneously.
 Shared-data
semaphore chopstick[5];
Initialization
chopstick[5]={1,1,1,1,1}.

Figure 2.21 Situation of dining philosophers Figure 2.22 The structure of philosopher

• Disadvantage:
2.6.3.1 Deadlock may occur if all 5 philosophers become hungry simultaneously and grab
their left chopstick. When each philosopher tries to grab her right chopstick, she will be
delayed forever.
• Three possible remedies to the deadlock problem:
1) Allow at most 4 philosophers to be sitting simultaneously at the table.
2) Allow a philosopher to pick up her chopsticks only if both chopsticks are available.
3) Use an asymmetric solution; i.e. an odd philosopher picks up first her left chopstick and then
her right chopstick, whereas an even philosopher picks up her right chopstick and then her left
chopstick.
 MODULE 3 conti…: DEADLOCKS

3.1Deadlocks
• Deadlock is a situation where a set of processes are blocked because each process is
→ holding a resource and
→ waiting for another resource held by some other process.
• Real life example:
When 2 trains are coming toward each other on same track and there is only one track, none of
the trains can move once they are in front of each other.
• Similar situation occurs in operating systems when there are two or more processes hold some
resources and wait for resources held by other(s).
• Here is an example of a situation where deadlock can occur (Figure 3.1).

Figure 3.1 Deadlock Situation

3.2System Model
• A system consist of finite number of resources. (For ex: memory, printers, CPUs).
• These resources are distributed among number of processes.
• A process must
→ request a resource before using it and
→ release the resource after using it.
• The process can request any number of resources to carry out a given task.
• The total number of resource requested must not exceed the total number of resources available.
• In normal operation, a process must perform following tasks in sequence:
1) Request
 If the request cannot be granted immediately (for ex: the resource is being used by another
process), then the requesting-process must wait for acquiring the resource.
 For example: open( ), malloc( ), new( ), and request( )
2) Use
 The process uses the resource.
 For example: prints to the printer or reads from the file.
3) Release
 The process releases the resource.
 So that, the resource becomes available for other processes.
 For example: close( ), free( ), delete( ), and release( ).
• A set of processes is deadlocked when every process in the set is waiting for a resource that is
currently allocated to another process in the set.
• Deadlock may involve different types of resources.
• As shown in figure 3.2,
Both processes P1 & P2 need resources to continue
execution.
P1 requires additional resource R1 and is in
possession of resource R2.
P2 requires additional resource R2 and is in
possession of R1.
• Thus, neither process can continue.
• Multithread programs are good candidates for deadlock because
Figure 3.2 they compete for shared resources.
 3.3Deadlock Characterization
• In a deadlock, processes never finish executing, and system resources are tied up, preventing other
jobs from starting.

3.3.1 Necessary Conditions

• There are four conditions that are necessary to achieve deadlock:
1) Mutual Exclusion
 At least one resource must be held in a non-sharable mode.
 If any other process requests this resource, then the requesting-process must wait for the
resource to be released.
2) Hold and Wait
 A process must be simultaneously
→ holding at least one resource and
→ waiting to acquire additional resources held by the other process.
3) No Preemption
 Once a process is holding a resource ( i.e. once its request has been granted ), then that
resource cannot be taken away from that process until the process voluntarily releases it.
4) Circular Wait
 A set of processes { P0, P1, P2, . . ., PN } must exist such that
P0 is waiting for a resource that is held by P1
P1 is waiting for a resource that is held by P2, and so on

3.3.2 Resource-Allocation-Graph
• The resource-allocation-graph (RAG) is a directed graph that can be used to describe the deadlock
situation.
• RAG consists of a
→ set of vertices (V) and
→ set of edges (E).
• V is divided into two types of nodes
1) P={P1,P2…....Pn} i.e., set consisting of all active processes in the system.
2) R={R1,R2….....Rn} i.e., set consisting of all resource types in the system.
• E is divided into two types of edges:
1) Request Edge
 A directed-edge Pi → Rj is called a request edge.
 Pi → Rj indicates that process Pi has requested a resource Rj.
2) Assignment Edge
 A directed-edge Rj → Pi is called an assignment edge.
 Rj → Pi indicates that a resource Rj has been allocated to process Pi.
• Suppose that process Pi requests resource Rj.
Here, the request for Rj from Pi can be granted only if the converting request-edge to
assignment-edge do not form a cycle in the resource-allocation graph.
• Pictorially,
→ We represent each process Pi as a circle.
→ We represent each resource-type Rj as a rectangle.
• As shown in below figures, the RAG illustrates the following 3 situation (Figure 3.3):
1) RAG with a deadlock
2) RAG with a cycle and deadlock
3) RAG with a cycle but no deadlock
 (a) Resource allocation Graph (b) With a deadlock (c) with cycle but no deadlock
Figure 3.3 Resource allocation graphs

Conclusion:
1) If a graph contains no cycles, then the system is not deadlocked.
2) If the graph contains a cycle then a deadlock may exist.
Therefore, a cycle means deadlock is possible, but not necessarily present.

3.4Methods for Handling Deadlocks

• There are three ways of handling deadlocks:
1) Deadlock prevention or avoidance - Do not allow the system to get into a deadlocked state.
2) Deadlock detection and recovery - Abort a process or preempt some resources when
deadlocks are detected.
3) Ignore the problem all together - If deadlocks only occur once a year or so, it may be better
to simply let them happen and reboot the system.
• In order to avoid deadlocks, the system must have additional information about all processes.
• In particular, the system must know what resources a process will or may request in the future.
• Deadlock detection is fairly straightforward, but deadlock recovery requires either aborting processes
or preempting resources.
• If deadlocks are neither prevented nor detected, then when a deadlock occurs the system will
gradually slow down.

3.5Deadlock-Prevention
• Deadlocks can be eliminated by preventing at least one of the four required conditions:
1) Mutual exclusion
2) Hold-and-wait
3) No preemption
4) Circular-wait.

3.5.1 Mutual Exclusion

• This condition must hold for non-sharable resources.
• For example:
A printer cannot be simultaneously shared by several processes.
• On the other hand, shared resources do not lead to deadlocks.
• For example:
Simultaneous access can be granted for read-only file.
• A process never waits for accessing a sharable resource.
• In general, we cannot prevent deadlocks by denying the mutual-exclusion condition because some
resources are non-sharable by default.

3.5.2 Hold and Wait

• To prevent this condition:
 The processes must be prevented from holding one or more resources while simultaneously
waiting for one or more other resources.
• There are several solutions to this problem.
• For example:
Consider a process that
→ copies the data from a tape drive to the disk
→ sorts the file and
→ then prints the results to a printer.
Protocol-1
 Each process must be allocated with all of its resources before it begins execution.
 All the resources (tape drive, disk files and printer) are allocated to the process at the
beginning.
Protocol-2
 A process must request a resource only when the process has none.
 Initially, the process is allocated with tape drive and disk file.
 The process performs the required operation and releases both tape drive and disk file.
 Then, the process is again allocated with disk file and the printer
 Again, the process performs the required operation & releases both disk file and the printer.
• Disadvantages of above 2 methods:
1) Resource utilization may be low, since resources may be allocated but unused for a long
period.
2) Starvation is possible.
3.5.3 No Preemption
• To prevent this condition: the resources must be preempted.
• There are several solutions to this problem.
Protocol-1
• If a process is holding some resources and requests another resource that cannot be immediately
allocated to it, then all resources currently being held are preempted.
• The preempted resources are added to the list of resources for which the process is waiting.
• The process will be restarted only when it regains the old resources and the new resources that it is
requesting.
Protocol-2
• When a process request resources, we check whether they are available or not.
If (resources are available)
then
{
allocate resources to the process
}
else
{
If (resources are allocated to waiting process)
then
{
preempt the resources from the waiting process
allocate the resources to the requesting-process
the requesting-process must wait
}

}
• These 2 protocols may be applicable for resources whose states are easily saved and restored, such
as registers and memory.
• But, these 2 protocols are generally not applicable to other devices such as printers and tape drives.

3.5.4 Circular-Wait
• Deadlock can be prevented by using the following 2 protocol:
Protocol-1
 Assign numbers all resources.
 Require the processes to request resources only in increasing/decreasing order.
Protocol-2
  Require that whenever a process requests a resource, it has released resources with a lower
number.
• One big challenge in this scheme is determining the relative ordering of the different resources.

3.6Deadlock Avoidance
• The general idea behind deadlock avoidance is to prevent deadlocks from ever happening.
• Deadlock-avoidance algorithm
→ requires more information about each process, and
→ tends to lead to low device utilization.
• For example:
1) In simple algorithms, the scheduler only needs to know the maximum number of each
resource that a process might potentially use.
2) In complex algorithms, the scheduler can also take advantage of the schedule of exactly
what resources may be needed in what order.
• A deadlock-avoidance algorithm dynamically examines the resources allocation state to ensure that
a circular-wait condition never exists.
• The resource-allocation state is defined by
→ the number of available and allocated resources and
→ the maximum demand of each process.

3.6.1 Safe State

• A state is safe if the system can allocate all resources requested by all processes without entering a
deadlock state.
• A state is safe if there exists a safe sequence of processes {P0, P1, P2, ..., PN} such that
the requests of each process(Pi) can be satisfied by the currently available resources.
• If a safe sequence does not exist, then the system is in an unsafe state, which may lead to deadlock.
• All safe states are deadlock free, but not all unsafe states lead to deadlocks. (Figure 3.4).

Figure 3.4 Safe, unsafe, and deadlock state spaces

3.6.2 Resource-Allocation-Graph Algorithm
• If resource categories have only single instances of their resources, then deadlock states can be
detected by cycles in the resource-allocation graphs.
• In this case, unsafe states can be recognized and avoided by augmenting the resource-allocation
graph with claim edges (denoted by a dashed line).
• Claim edge Pi → Rj indicated that process Pi may request resource Rj at some time in future.
• The important steps are as below:
1) When a process Pi requests a resource Rj, the claim edge Pi → Rj is converted to a request
edge.
2) Similarly, when a resource Rj is released by the process Pi, the assignment edge Rj → Pi is
reconverted as claim edge Pi → Rj.
3) The request for Rj from Pi can be granted only if the converting request edge to assignment
edge do not form a cycle in the resource allocation graph.
• To apply this algorithm, each process Pi must know all its claims before it starts executing.
• Conclusion:
1) If no cycle exists, then the allocation of the resource will leave the system in a safe state.
2) If cycle is found, system is put into unsafe state and may cause a deadlock.
• For example: Consider a resource allocation graph shown in Figure 3.5(a).
 Suppose P2 requests R2.
 Though R2 is currently free, we cannot allocate it to P2 as this action will create a cycle in the
graph as shown in Figure 3.5(b).
 This cycle will indicate that the system is in unsafe state: because, if P1 requests R2 and P2
requests R1 later, a deadlock will occur.

(a) For deadlock avoidance (b) an unsafe state

Figure 3.5 Resource Allocation graphs

• Problem:
The resource-allocation graph algorithm is not applicable when there are multiple instances
for each resource.
• Solution:
Use banker's algorithm.
 3.6.3 Banker's Algorithm
• This algorithm is applicable to the system with multiple instances of each resource types.
• However, this algorithm is less efficient then the resource-allocation-graph algorithm.
• When a process starts up, it must declare the maximum number of resources that it may need.
• This number may not exceed the total number of resources in the system.
• When a request is made, the system determines whether granting the request would leave the
system in a safe state.
• If the system in a safe state,
the resources are allocated;
else
the process must wait until some other process releases enough resources.
• Assumptions:
Let n = number of processes in the system
Let m = number of resources types.
• Following data structures are used to implement the banker’s algorithm.
1) Available [m]
 This vector indicates the no. of available resources of each type.
 If Available[j]=k, then k instances of resource type Rj is available.
2) Max [n][m]
 This matrix indicates the maximum demand of each process of each resource.
 If Max[i,j]=k, then process Pi may request at most k instances of resource type Rj.
3) Allocation [n][m]
 This matrix indicates no. of resources currently allocated to each process.
 If Allocation[i,j]=k, then Pi is currently allocated k instances of Rj.
4) Need [n][m]
 This matrix indicates the remaining resources need of each process.
 If Need[i,j]=k, then Pi may need k more instances of resource Rj to complete its task.
 So, Need[i,j] = Max[i,j] - Allocation[i]
• The Banker’s algorithm has two parts: 1) Safety Algorithm
2) Resource – Request Algorithm

3.6.3.1 Safety Algorithm

• This algorithm is used for finding out whether a system is in safe state or not.
• Assumptions:
Work is a working copy of the available resources, which will be modified during the analysis.
Finish is a vector of boolean values indicating whether a particular process can finish.

Step 1:
Let Work and Finish be two vectors of length m and n respectively.
Initialize:
Work = Available
Finish[i] = false for i=1,2,3,…….n
Step 2:
Find an index(i) such that both
a) Finish[i] = false
b) Need i <= Work.
If no such i exist, then go to step 4
Step 3:
Set:
Work = Work + Allocation(i)
Finish[i] = true
Go to step 2
Step 4:
If Finish[i] = true for all i, then the system is in safe state.
 3.6.3.2 Resource-Request Algorithm
• This algorithm determines if a new request is safe, and grants it only if it is safe to do so.
• When a request is made ( that does not exceed currently available resources ), pretend it has been
granted, and then see if the resulting state is a safe one. If so, grant the request, and if not, deny the
request.
• Let Request(i) be the request vector of process Pi.
• If Request(i)[j]=k, then process Pi wants K instances of the resource type Rj.

Step 1:
If Request(i) <= Need(i)
then
go to step 2
else
raise an error condition, since the process has exceeded its maximum claim.
Step 2:
If Request(i) <= Available
then
go to step 3
else
Pi must wait, since the resources are not available.
Step 3:
If the system want to allocate the requested resources to process Pi then modify the
state as follows:
Available = Available – Request(i)
Allocation(i) = Allocation(i) + Request(i)
Need(i) = Need(i) – Request(i)
Step 4:
If the resulting resource-allocation state is safe,
then i) transaction is complete and
ii) Pi is allocated its resources.
Step 5:
If the new state is unsafe,
then i) Pi must wait for Request(i) and
ii) old resource-allocation state is restored.
 3.6.3.3 An Illustrative Example
Question: Consider the following snapshot of a system:
Allocation Max Available
A B C A B C A B C
P 0 1 0 7 5 3 3 3 2
0
P 2 0 0 3 2 2
1
P 3 0 3 9 0 2
2
P 2 1 1 2 2 2
3
P 0 0 2 4 3 3
4

Answer the following questions using Banker's algorithm.

i) What is the content of the matrix need?
ii) Is the system in a safe state?
iii) If a request from process P1 arrives for (1 0 2) can the request be granted immediately?

Solution (i):
• The content of the matrix Need is given by
Need = Max - Allocation
• So, the content of Need Matrix is:
Need
A B C
P0 7 4 3
P1 1 2 2
P2 6 0 0
P3 0 1 1
P4 4 3 1
Solution (ii):
• Applying the Safety algorithm on the given system,
Step 1: Initialization
Work = Available i.e. Work =3 3 2
……P0………P1……..P2……..P3……P4…..
Finish = | false | false | false | false | false |

Step 2: For i=0

Finish[P0] = false and Need[P0]<=Work i.e. (7 4 3)<=(3 3 2)  false So
P0 must wait.

Step 2: For i=1

Finish[P1] = false and Need[P1]<=Work i.e. (1 2 2)<=(3 3 2)  true So
P1 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P1] =(3 3 2)+(2 0 0)=(5 3 2)
…….P0………P1…….P2…….P3… P4……
Finish = | false | true | false | false | false |

Step 2: For i=2

Finish[P2] = false and Need[P2]<=Work i.e. (6 0 0)<=(5 3 2)  false So
P2 must wait.

Step 2: For i=3

Finish[P3] = false and Need[P3]<=Work i.e. (0 1 1)<=(5 3 2)  true So
P3 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P3] = (5 3 2)+(2 1 1)=(7 4 3)
……P0………P1……P2………..P3…….P4….
 Finish = | false | true | false | true | false |

Step 2: For i=4

Finish[P4] = false and Need[P4]<=Work i.e. (4 3 1)<=(7 4 3)  true So
P4 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P4] =(7 4 3)+(0 0 2)=(7 4 5)
……P0………P1……P2………P3…….P4…..
Finish= | false | true | false | true | true |

Step 2: For i=0

Finish[P0] = false and Need[P0]<=Work i.e. (7 4 3)<=(7 4 5)  true So
P0 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P0] =(7 4 5)+(0 1 0)=(7 5 5)
…..P0…….P1………P2…….P3…….P4….
Finish= | true | true | false | true | true |

Step 2: For i=2

Finish[P2] = false and Need[P2]<=Work i.e. (6 0 0) <=(7 5 5)  true So
P2 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P2] =(7 5 5)+(3 0 2)=(10 5 7)
…..P0…….P1………P2…….P3…….P4….
Finish= | true | true | true | true | true |

Step 4: Finish[Pi] = true for 0<=i<=4

Hence, the system is currently in a safe state.
The safe sequence is <P1, P3, P4, P0, P2>.
Conclusion: Yes, the system is currently in a safe state.

Solution (iii): P1 requests (1 0 2) i.e. Request[P1]=1 0 2

• To decide whether the request is granted, we use Resource Request algorithm.
Step 1: Request[P1]<=Need[P1] i.e. (1 0 2)<=(1 2 2)  true.
Step 2: Request[P1]<=Available i.e. (1 0 2)<=(3 3 2)  true.
Step 3: Available = Available – Request[P1] = (3 3 2) - (1 0 2)= (2 3 0)
Allocation[P1] = Allocation[P1] + Request[P1] = (2 0 0) + (1 0 2)= (3 0 2)
Need[P1] = Need[P1] – Request[P1] = (1 2 2) - (1 0 2)= (0 2 0)
• We arrive at the following new system state:

Allocation Max Available

A B C A B C A B C
P 0 1 0 7 5 3 2 3 0
0
• The content of the matrix P 3 0 2 3 2 2 Need is given by
Need = Max - 1 Allocation
• So, the content of Need P 3 0 2 9 0 2 Matrix is:
2
P 2 1 1 2 2 2
3
P 0 0 2 4 3 3
4
Need
A B C
P0 7 4 3
P1 0 2 0
 • To determine whether this new system state is safe, we again execute Safety algorithm.
Step 1: Initialization
Here, m=3, n=5
Work = Available i.e. Work =2 3 0
…..P0………P1………P2……….P3…….P4….
Finish = | false | false | false | false | false

Step 2: For i=0

Finish[P0] = false and Need[P0]<=Work i.e. (7 4 3)<=(2 3 0)  false So
P0 must wait.

Step 2: For i=1

Finish[P1] = false and Need[P1]<=Work i.e. (0 2 0)<=(2 3 0)  true So
P1 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P1] =(2 3 0)+(3 0 2)=(5 3 2)
……P0………P1…….P2………P3…….P4……
Finish = | false | true | false | false | false |

Step 2: For i=2

Finish[P2] = false and Need[P2]<=Work i.e. (6 0 0) <=(5 3 2)  false So
P2 must wait.

Step 2: For i=3

Finish[P3] = false and Need[P3]<=Work i.e. (0 1 1)<=(5 3 2)  true So
P3 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P3] = (5 3 2)+(2 1 1)=(7 4 3)
…..P0……….P1……P2……..P3…….P4…...
Finish = | false | true | false | true | false |

Step 2: For i=4

Finish[P4] = false and Need[P4]<=Work i.e. (4 3 1)<=(7 4 3)  true So
P4 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P4] =(7 4 3)+(0 0 2)=(7 4 5)
……P0………P1……P2………P3…….P4….
Finish = | false | true | false | true | true |

Step 2: For i=0

Finish[P0] = false and Need[P0]<=Work i.e. (7 4 3)<=(7 4 5)  true So
P0 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P0] =(7 4 5)+(0 1 0)=(7 5 5)
……P0…….P1………P2……P3. .P4….
Finish = | true | true | false | true | true |

Step 2: For i=2

Finish[P2] = false and Need[P2]<=Work i.e. (6 0 0) <=(7 5 5)  true So
P2 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P2] =(7 5 5)+(3 0 2)=(10 5 7)
…..P0…….P1………P2…….P3…….P4….
Finish= | true | true | true | true | true |

Step 4: Finish[Pi] = true for 0<=i<=4

Hence, the system is in a safe state.
The safe sequence is <P1, P3, P4, P0, P2>.
Conclusion: Since the system is in safe sate, the request can be granted.
 3.7Deadlock Detection
• If a system does not use either deadlock-prevention or deadlock-avoidance algorithm then a
deadlock may occur.
• In this environment, the system must provide
1) An algorithm to examine the system-state to determine whether a deadlock has occurred.
2) An algorithm to recover from the deadlock.

3.7.1 Single Instance of Each Resource Type

• If all the resources have only a single instance, then deadlock detection-algorithm can be defined
using a wait-for-graph.
• The wait-for-graph is applicable to only a single instance of a resource type.
• A wait-for-graph (WAG) is a variation of the resource-allocation-graph.
• The wait-for-graph can be obtained from the resource-allocation-graph by
→ removing the resource nodes and
→ collapsing the appropriate edges.
• An edge from Pi to Pj implies that process Pi is waiting for process Pj to release a resource that Pi
needs.
• An edge Pi → Pj exists if and only if the corresponding graph contains two edges
1) Pi → Rq and
2) Rq → Pj.
• For example:
Consider resource-allocation-graph shown in Figure 3.6
Corresponding wait-for-graph is shown in Figure 3.7.

Figure 3.6 Resource-allocation-graph Figure 3.7 Corresponding wait-for-graph.

• A deadlock exists in the system if and only if the wait-for-graph contains a cycle.
• To detect deadlocks, the system needs to
→ maintain the wait-for-graph and
→ periodically execute an algorithm that searches for a cycle in the graph.
 3.7.2 Several Instances of a Resource Type
• The wait-for-graph is applicable to only a single instance of a resource type.
• Problem: However, the wait-for-graph is not applicable to a multiple instance of a resource type.
• Solution: The following detection-algorithm can be used for a multiple instance of a resource type.
• Assumptions:
Let ‘n’ be the number of processes in the system
Let ‘m’ be the number of resources types.
• Following data structures are used to implement this algorithm.
1) Available [m]
 This vector indicates the no. of available resources of each type.
 If Available[j]=k, then k instances of resource type Rj is available.
2) Allocation [n][m]
 This matrix indicates no. of resources currently allocated to each process.
 If Allocation[i,j]=k, then Pi is currently allocated k instances of Rj.
3) Request [n][m]
 This matrix indicates the current request of each process.
 If Request [i, j] = k, then process Pi is requesting k more instances of resource type Rj.
Step 1:
Let Work and Finish be vectors of length m and n respectively.
a) Initialize Work = Available
b) For i=0,1,2…......n
if Allocation(i) != 0
then
Finish[i] = false;
else
Finish[i] = true;
Step 2:
Find an index(i) such that both
a) Finish[i] = false
b) Request(i) <= Work.
If no such i exist, goto step 4.
Step 3:
Set:
Work = Work + Allocation(i)
Finish[i] = true
Go to step 2.
Step 4:
If Finish[i] = false for some i where 0 < i < n, then the system is in a deadlock state.

3.7.3 Detection-Algorithm Usage

• The detection-algorithm must be executed based on following factors:
1) The frequency of occurrence of a deadlock. 2) The no. of processes affected by the deadlock.
• If deadlocks occur frequently, then the detection-algorithm should be executed frequently.
• Resources allocated to deadlocked-processes will be idle until the deadlock is broken.
• Problem:
Deadlock occurs only when some processes make a request that cannot be granted immediately.
• Solution 1:
 The deadlock-algorithm must be executed whenever a request for allocation cannot be
granted immediately.
 In this case, we can identify
→ set of deadlocked-processes and
→ specific process causing the deadlock.
• Solution 2:
 The deadlock-algorithm must be executed in periodic intervals.
 For example:
→ once in an hour
→ whenever CPU utilization drops below certain threshold
 3.8Recovery from deadlock
• Three approaches to recovery from deadlock:
1) Inform the system-operator for manual intervention.
2) Terminate one or more deadlocked-processes.
3) Preempt(or Block) some resources.

3.8.1 Process Termination

• Two methods to remove deadlocks:
1) Terminate all deadlocked-processes.
 This method will definitely break the deadlock-cycle.
 However, this method incurs great expense. This is because
→ Deadlocked-processes might have computed for a long time.
→ Results of these partial computations must be discarded.
→ Probably, the results must be re-computed later.
2) Terminate one process at a time until the deadlock-cycle is eliminated.
 This method incurs large overhead. This is because
after each process is aborted,
deadlock-algorithm must be executed to determine if any other process is still deadlocked
• For process termination, following factors need to be considered:
1) The priority of process.
2) The time taken by the process for computation & the required time for complete execution.
3) The no. of resources used by the process.
4) The no. of extra resources required by the process for complete execution.
5) The no. of processes that need to be terminated for deadlock-free execution.
6) The process is interactive or batch.

3.8.2 Resource Preemption

• Some resources are taken from one or more deadlocked-processes.
• These resources are given to other processes until the deadlock-cycle is broken.
• Three issues need to be considered:
1) Selecting a victim
 Which resources/processes are to be pre-empted (or blocked)?
 The order of pre-emption must be determined to minimize cost.
 Cost factors includes
1. The time taken by deadlocked-process for computation.
2. The no. of resources used by deadlocked-process.
2) Rollback
 If a resource is taken from a process, the process cannot continue its normal execution.
 In this case, the process must be rolled-back to break the deadlock.
 This method requires the system to keep more info. about the state of all running processes.
3) Starvation
 Problem: In a system where victim-selection is based on cost-factors, the same process may
be always picked as a victim.
 As a result, this process never completes its designated task.
 Solution: Ensure a process is picked as a victim only a (small) finite number of times.
 Exercise Problems

1) Consider the following snapshot of a system:

Allocation Max Available
A B C A B C A B C
P 0 0 2 0 0 4 1 0 2
0
P 1 0 0 2 0 1
1
P 1 3 5 1 3 7
2
P 6 3 2 8 4 2
3
P 1 4 3 1 5 7
4

Answer the following questions using Banker's algorithm:

i) What is the content of the matrix need?
ii) Is the system in a safe state?
iii) If a request from process P2 arrives for (0 0 2) can the request be granted immediately?

Solution (i):
• The content of the matrix Need is given by
Need = Max - Allocation
• So, the content of Need Matrix is:
Need
A B C
P0 0 0 2
P1 1 0 1
P2 0 0 2
P3 2 1 0
P4 0 1 4
Solution (ii):
• Applying the Safety algorithm on the given system,
Step 1: Initialization
Work = Available i.e. Work =1 0 2
……P0………P1……..P2……..P3……P4…..
Finish = | false | false | false | false | false |

Step 2: For i=0

Finish[P0] = false and Need[P0]<=Work i.e. (0 0 2)<=(1 0 2)  true So
P0 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P0] =(1 0 2)+(0 0 2)=(1 0 4)
……P0………P1…….P2…...P3……P4….
Finish = | true | false | false | false | false |

Step 2: For i=1

Finish[P1] = false and Need[P1]<=Work i.e. (1 0 1)<=(1 0 4)  true So
P1 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P1] =(1 0 4)+(1 0 0)=(2 0 4)
…….P0…….P1……P2………P3… P4…
Finish = | true | true | false | false | false |

Step 2: For i=2

Finish[P2] = false and Need[P2]<=Work i.e. (0 0 2)<=(2 0 4)  true So
P2 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P2] =(2 0 4)+(1 3 5)=(3 3 9)
……P0…….P1…….P2…. .P3……P4….
Finish = | true | true | true | false | false |
 Step 2: For i=3
Finish[P3] = false and Need[P3]<=Work i.e. (2 1 0)<=(3 3 9)  true So
P3 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P3] = (3 3 9)+(6 3 2)=(9 6 11)
…..P0……..P1……P2……..P3…….P4….
Finish = | true | true | true | true | false |

Step 2: For i=4

Finish[P4] = false and Need[P4]<=Work i.e. (0 1 4)<=(9 6 11)  true So
P4 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P4] =(9 6 11)+(1 4 3)=(10 10 14)
……P0…….P1……..P2……P3…….P4….
Finish = | true | true | true | true | true |

Step 4: Finish[Pi] = true for 0<=i<=4

Hence, the system is currently in a safe state.
The safe sequence is <P0, P1, P2, P3, P4>.
Conclusion: Yes, the system is currently in a safe state.

Solution (iii): P2 requests (0 0 2) i.e. Request[P2]=0 0 2

• To decide whether the request is granted, we use Resource Request algorithm.
Step 1: Request[P2]<=Need[P2] i.e. (0 0 2) <= (1 3 7)  true.
Step 2: Request[P2]<=Available i.e. (0 0 2) <= (1 0 2)  true.
Step 3: Available = Available – Request[P2] = (1 0 2) - (0 0 2)= (1 0 0)
Allocation[P2] = Allocation[P2] + Request[P2] = (1 3 5) + (0 0 2)= (1 3 7)
Need[P2] = Need[P2] – Request[P2] = (0 0 2) - (0 0 2)= (0 0 0)
• We arrive at the following new system state:
Allocation Max Available
A B C A B C A B C
P 0 0 2 0 0 4 1 0 0
0
P 1 0 0 2 0 1
1
P 1 3 7 1 3 7
2
P 6 3 2 8 4 2
3
P 1 4 3 1 5 7
4

• The content of the matrix Need is given by

Need = Max - Allocation
• So, the content of Need Matrix is:
Need
A B C
P0 0 0 2
P1 1 0 1
P2 0 0 0
P3 2 1 0
P4 0 1 4
• To determine whether this new system state is safe, we again execute Safety algorithm.
Step 1: Initialization
Work = Available i.e. Work =2 3 0
....P0………P1…….P2……….P3…..P4…..
Finish = | false | false | false | false | false |

Step 2: For i=0

Finish[P0] = false and Need[P0]<=Work i.e. (0 0 2)<=(2 3 0)  false So
 P0 must wait.

Step 2: For i=1

Finish[P1] = false and Need[P1]<=Work i.e. (1 0 1)<=(2 3 0)  false So
P1 must wait.

Step 2: For i=2

Finish[P2] = false and Need[P2]<=Work i.e. (0 0 0)<=(2 3 0)  true So
P2 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P2] =(1 0 0)+(1 3 7)=(2 3 7)
.....P0…….P1……..P2…..P3.......P4….
Finish = | false | false | true | false | false |

Step 2: For i=3

Finish[P3] = false and Need[P3]<=Work i.e. (2 1 0)<=(2 3 7)  true So
P3 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P3] = (2 3 7)+(6 3 2)=(8 6 9)
....P0……….P1…….P2…….P3….P4…
Finish = | false | false | true | true | false |

Step 2: For i=4

Finish[P4] = false and Need[P4]<=Work i.e. (0 1 4)<=(8 6 9)  true So
P4 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P4] =(8 6 9)+(0 1 4)=(8 7 13)
....P0……….P1…….P2…….P3….P4…
Finish = | false | false | true | true | true |

Step 2: For i=0

Finish[P0] = false and Need[P0]<=Work i.e. (0 0 2)<=(8 7 13)  true
So P0 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P0] =(8 7 13)+(0 0 2)=(8 7 15)
....P0……….P1…….P2…….P3…….P4…
Finish = | true | false | true | true | true |

Step 2: For i=1

Finish[P1] = false and Need[P1]<=Work i.e. (1 0 1)<=(8 7 15)  true
So P1 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P1] =(8 7 15)+(1 0 0)=(9 7 15)
....P0……….P1…….P2……P3…….P4…
Finish = | true | true | true | true | true |

Step 4: Finish[Pi] = true for 0<=i<=4

Hence, the system is in a safe state.
The safe sequence is <P2, P3, P4, P0, P1>.
Conclusion: Since the system is in safe sate, the request can be granted.
 [Type the document title]

2) For the following snapshot, find the safe sequence using Banker's algorithm:
The number of resource units is (A, B, C) which are (7, 7, 10) respectively.
Allocation Max Available
A B C A B C A B C
P1 2 2 3 3 6 8 7 7 10
P2 2 0 3 4 3 3
P3 1 2 4 3 4 4
Solution:
• The content of the matrix Need is given by
Need = Max - Allocation
• So, the content of Need Matrix is:
Need
A B C
P1 1 4 5
P2 2 3 0
P3 2 2 0
• Applying the Safety algorithm on the given system,
Step 1: Initialization
Here, m=3, n=3
Work = Available i.e. Work =7 7 10
….P1………..P2…. .P3…
Finish = | false | false | false |
Step 2: For i=1
Finish[P1] = false and Need[P1]<=Work i.e. (1 4 5)<=(7 7 10)  true So
P1 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P1] =(7 7 10)+(2 2 3)=(9 9 13)
……P1……P2……….P3….
Finish = | true | false | false |

Step 2: For i=2

Finish[P2] = false and Need[P2]<=Work i.e. (2 3 0) <=(9 9 13)  true
So P2 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P2] =(9 9 13)+(2 0 3)=(11 9 16)
…..P1……P2…….P3……
Finish = | true | true | false |

Step 2: For i=3

Finish[P3] = false and Need[P3]<=Work i.e. (2 2 0)<=(11 9 16)  true So
P3 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P3] = (11 9 16)+(1 2 4)=(12 11 20)
……P1……P2…….P3….
Finish = | true | true | true |

Step 4: Finish[Pi] = true for 1<=i<=3

Hence, the system is currently in a safe state.
The safe sequence is <P1, P2, P3>.
Conclusion: Yes, the system is currently in a safe state.

Page 20
 [Type the document title]

3) Consider the following snapshot of resource-allocation at time t1.

Allocation Max Available
A B C A B C A B C
P 0 1 0 0 0 0 0 0 0
0
P 2 0 0 2 0 2
1
P 3 0 3 0 0 0
2
P 2 1 1 1 0 0
3
P 0 0 2 0 0 2
4
i) What is the content of the matrix need?
ii) Show that the system is not deadlock by generating one safe sequence
iii) At instance t, P2 makes one additional for instance of type C. Show that the system is deadlocked if
the request is granted. Write down deadlocked-processes.

Solution (i):
• The content of the matrix Need is given by
Need = Max - Allocation
• So, the content of Need Matrix is:
Need
A B C
P0 0 0 0
P1 0 0 2
P2 0 0 0
P3 0 0 0
P4 0 0 0
Solution (ii):
• Applying the Safety algorithm on the given system,
Step 1: Initialization
Work = Available i.e. Work =0 0 0
……P0………P1…….P2…. P3……P4…
Finish = | false | false | false | false | false |

Step 2: For i=0

Finish[P0] = false and Need[P0]<=Work i.e. (0 0 0)<=(0 0 0)  true So
P0 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P0] =(0 0 0)+(0 1 0)=(0 1 0)
…..P0………P1…….P2…….P3…P4…
Finish = | true | false | false | false | false |

Step 2: For i=1

Finish[P1] = false and Need[P1]<=Work i.e. (0 0 2)<=(0 1 0)  false So
P1 must wait.

Step 2: For i=2

Finish[P2] = false and Need[P2]<=Work i.e. (0 0 0) <=(0 1 0)  true So
P2 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P2] =(0 1 0)+(3 0 3)=(5 1 3)
....P0……P1…….P2……….P3…….P4…
Finish = | true | false | true | false | false |

Step 2: For i=3

Finish[P3] = false and Need[P3]<=Work i.e. (0 0 0)<=(5 1 3)  true So
P3 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P3] = (5 1 3)+(2 1 1)=(5 2 4)

Page 21
[Type the document title]

…P0……P1………P2…….P3…….P4…
Finish = | true | false | true | true | false |

Step 2: For i=4

Finish[P4] = false and Need[P4]<=Work i.e. (0 0 0)<=(5 2 4)  true So
P4 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P4] =(5 2 4)+(0 0 2)=(5 2 6)
....P0……P1…….P2…….P3…….P4….
Finish = | true | false | true | true | true |

Step 2: For i=1

Finish[P1] = false and Need[P1]<=Work i.e. (0 0 2)<=(5 2 6)  true So
P0 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P1] =(5 2 6)+(2 0 0)=(7 2 6)
...P0…….P1……..P2…….P3…….P4…
Finish= | true | true | true | true | true |

Step 4: Finish[Pi] = true for 0<=i<=4

Hence, the system is currently in a safe state.
The safe sequence is <P0, P2, P3, P4, P1>.
Conclusion: Yes, the system is currently in a safe state. Hence there is no deadlock in the system.

Solution (iii): P2 requests (0 0 1) i.e. Request[P1]=0 0 1

• To decide whether the request is granted, we use Resource Request algorithm.
Step 1: Request[P1]<=Need[P1] i.e. (0 0 1)<=(0 0 2)  true.
Step 2: Request[P1]<=Available i.e. (0 0 1)<=(0 0 0)  false.
Conclusion: Since Request[P1]>Available, we cannot process this request.
Since P2 will be in waiting state, deadlock occurs in the system.

4) For the given snapshot :

Allocation Max Available

Page 22
 [Type the document title]

A B C D A B C D A B C D
P1 0 0 1 2 0 0 1 2 1 5 2 0
P2 1 0 0 0 1 7 5 0
P3 1 3 5 4 2 3 5 6
P4 0 6 3 2 0 6 5 2
P5 0 0 1 4 0 6 5 6

Using Banker's algorithm:

i) What is the need matrix content?
ii) Is the system in safe state?
iii) If a request from process P2(0,4,2,0) arrives, can it be granted?

Solution (i):
• The content of the matrix Need is given by
Need = Max - Allocation
• So, the content of Need Matrix is:
Need
A B C D
P 0 0 0 0
1
P 0 7 5 2
2
P 1 0 0 2
3
P 0 0 2 0
4
P 0 6 4 2
5
Solution (ii):
• Applying the Safety algorithm on the given system,
Step 1: Initialization
Work = Available i.e. Work =1 5 2 0
....P1………P2…….P3……….P4…..P5…..
Finish = | false | false | false | false | false |

Step 2: For i=1

Finish[P1] = false and Need[P1]<=Work i.e. (0 0 0 0)<=(1 5 2 0)  true So
P1 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P1] =(1 5 2 0)+(0 0 1 2)=(1 5 3 2)
....P1………P2…….P3…….P4….P5…
Finish = | true | false | false | false | false |

Step 2: For i=2

Finish[P2] = false and Need[P2]<=Work i.e. (0 7 5 2)<=(1 5 3 2)  false
So P2 must wait.

Step 2: For i=3

Finish[P3] = false and Need[P3]<=Work i.e. (1 0 0 2)<=(1 5 3 2)  true So
P3 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P3] = (1 5 3 2)+(1 3 5 4)=(2 8 8 6)
....P1………P2…….P3…….P4….P5…
Finish = | true | false | true | false | false |

Step 2: For i=4

Finish[P4] = false and Need[P4]<=Work i.e. (0 0 2 0)<=(2 8 8 6)  true So
P4 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P4] =(2 8 8 6)+(0 6 3 2)=(2 14 11 8)
....P1………P2…….P3…….P4…….P5…
Finish = | true | false | true | true | false |

Page 23
[Type the document title]

Step 2: For i=5

Finish[P5] = false and Need[P5]<=Work i.e. (0 6 4 2)<=(2 14 11 8)  true So
P5 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P5] =(2 14 11 8)+(0 0 1 4)=(2 14 12 12)
....P1………P2…….P3…….P4……P5…
Finish = | true | false | true | true | true |

Step 2: For i=2

Finish[P2] = false and Need[P2]<=Work i.e. (0 7 5 2) <=(2 14 12 12)  true So
P2 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P2] =(2 14 12 12)+(1 0 0 0)=( 3 14 12 12)
....P1………P2…….P3…….P4……P5…
Finish = | true | true | true | true | true |

Step 4: Finish[Pi] = true for 1<=i<=5

Hence, the system is currently in a safe state.
The safe sequence is <P1, P3, P4, P5, P2>.
Conclusion: Yes, the system is currently in a safe state.

Solution (iii): P2 requests (0 4 2 0) i.e. Request[P2]= 0 4 2 0

• To decide whether the request is granted, we use Resource Request algorithm.
Step 1: Request[P2]<=Need[P2] i.e. (0 4 2 0) <= (0 7 5 2)  true. Step 2:
Request[P2]<=Available i.e. (0 4 2 0) <= (1 5 2 0)  true.
Step 3: Available = Available – Request[P2] = (1 5 2 0) - (0 4 2 0)= (1 1 0 0)
Allocation[P2] = Allocation[P2] + Request[P2] = (1 0 0 0) + (0 4 2 0)= (1 4 2 0)
Need[P2] = Need[P2] – Request[P2] = (0 7 5 2) - (0 4 2 0)= (0 3 3 2)
• We arrive at the following new system state

Allocation Max Available

A B C D A B C D A B C D
P1 0 0 1 2 0 0 1 2 1 1 0 0
P2 1 4 2 0 1 7 5 0
P3 1 3 5 4 2 3 5 6
P4 0 6 3 2 0 6 5 2
P5 0 0 1 4 0 6 5 6

• The content of the matrix Need is given by

Need = Max - Allocation
• So, the content of Need Matrix is:
Need
A B C D
P 0 0 0 0
1
P 0 3 3 2
2
P 1 0 0 2
3
P 0 0 2 0
4
P 0 6 4 2
5

• Applying the Safety algorithm on the given system,

Step 1: Initialization
Work = Available i.e. Work =1 1 0 0
....P1………P2…….P3………..P4… P5….
Finish = | false | false | false | false | false |

Step 2: For i=1

Finish[P1] = false and Need[P1]<=Work i.e. (0 0 0 0)<=(1 1 0 0)  true So

Page 24
[Type the document title]

P1 must be kept in safe sequence.

Step 3: Work = Work + Allocation[P1] =(1 1 0 0)+(0 0 1 2)=(1 1 1 2)
....P1………P2…….P3……...P4....P5…
Finish = | true | false | false | false | false |

Step 2: For i=2

Finish[P2] = false and Need[P2]<=Work i.e. (0 3 3 2)<=(1 1 1 2) false
So P2 must wait.

Step 2: For i=3

Finish[P3] = false and Need[P3]<=Work i.e. (1 0 0 2)<=(1 1 1 2)  true
So P3 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P3] = (1 1 1 2)+(1 3 5 4)=(2 4 6 6)
....P1……P2……….P3…….P4….P5….
Finish = | true | false | true | false | false |

Step 2: For i=4

Finish[P4] = false and Need[P4]<=Work i.e. (0 0 2 0)<=(2 4 6 6)  true So
P4 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P4] =(2 4 6 6)+(0 6 3 2)=(2 10 9 8)
....P1………P2…….P3…….P4…….P5….
Finish = | true | false | true | true | false |

Step 2: For i=5

Finish[P5] = false and Need[P5]<=Work i.e. (0 6 4 2)<=(2 10 9 8)  true So
P5 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P5] =(2 10 9 8)+(0 0 1 4)=(2 10 10 12)
....P1………P2…….P3…….P4……P5….
Finish = | true | false | true | true | true |

Step 2: For i=2

Finish[P2] = false and Need[P2]<=Work i.e. (0 3 3 2) <=(2 10 10 12)  true So
P2 must be kept in safe sequence.
Step 3: Work = Work + Allocation[P2] =(2 10 10 12)+(1 4 2 0)=(3 14 12 12)
....P1………P2……P3…….P4……P5….
Finish = | true | true | true | true | true |

Step 4: Finish[Pi] = true for 0<=i<=4

Hence, the system is currently in a safe state.
The safe sequence is <P1, P3, P4, P5, P2>.
Conclusion: Since the system is in safe sate, the request can be granted.

5) Consider a system containing ‘m’ resources of the same type being shared by ‘n’ processes.
Resources can be requested and released by processes only one at a time. Show that the system is
deadlock free if the following two conditions hold:
i) The maximum need of each process is between 1 and m resources
ii) The sum of all maximum needs is less than m+n.

Page 25
 [Type the document title]

Ans:
• Suppose N = Sum of all Needi
A = Sum of all Allocationi
M = Sum of all Maxi.
• Use contradiction to prove: Assume this system is not deadlock free.
• If there exists a deadlock state, then A=m because there's only one kind of resource and resources
can be requested and released only one at a time.
• From condition (ii), N+A = M<m+n
• So we get N+m <m +n.
• So we get N < n.
• It shows that at least one process i that Needi=0.
• From condition (i), Pi can release at least one resource.
• So, there are n-1 processes sharing ‘m’ resources now, condition (i) and (ii) still hold.
• Go on the argument, no process will wait permanently, so there's no deadlock.

6) Consider the traffic deadlock depicted in the figure given below, explain that the four necessary
conditions for dead lock indeed hold in this examples.

Ans:
• The four necessary conditions for a deadlock are:
1) Mutual exclusion
2) Hold-and-wait
3) No preemption and
4) Circular-wait.
• The mutual exclusion condition holds since only one car can occupy a space in the roadway.
• Hold-and-wait occurs where a car holds onto its place in the roadway while it waits to advance in the
roadway.
• A car cannot be removed (i.e. preempted) from its position in the roadway.
• Lastly, there is indeed a circular-wait as each car is waiting for a subsequent car to advance.
• The circular-wait condition is also easily observed from the graphic.

Page 26