5. deadlocks

7/24/2019 5. Deadlocks

1/43

Chapter 7Chapter 7

DeadlocksDeadlocks


2/43

7.2 Silberschatz, Galvin and Gagne 2005Operating System Concepts - 7thEdition, Feb 14, 2005

Chapter 7: DeadlocksChapter 7: Deadlocks

7.1 System Model 7.2 Deadlock Characterization

7.3 Methods for Handling Deadlocks

7.4 Deadlock Prevention

7.5 Deadlock Avoidance

7.6 Deadlock Detection

7.7 Recovery from Deadlock


3/43


Chapter ObjectivesChapter Objectives

To develop a description of deadlocks, which preventsets of concurrent processes from completing their tasks

To present a number of different methods for preventing,avoiding, or detecting deadlocks in a computer system


4/43

7.1 System Model7.1 System Model


5/43


The Deadlock ProblemThe Deadlock Problem

A deadlock consists of a set of blocked processes, eachholding a resource and waiting to acquire a resource held byanother process in the set

Example #1

A system has 2 disk drives

P1andP2each hold one disk drive and each needs the other one

Example #2

SemaphoresAandB, initialized to 1

P0 P1

wait (A); wait(B)wait (B); wait(A)


6/43


Bridge Crossing ExampleBridge Crossing Example

Traffic only in one direction

The resource is a one-lane bridge

If a deadlock occurs, it can be resolved if one car backs up(preempt resources and rollback)

Several cars may have to be backed up if a deadlock occurs

Starvation is possible


7/437.7 Silberschatz, Galvin and Gagne 2005Operating System Concepts - 7thEdition, Feb 14, 2005

System ModelSystem Model

Resource typesR1,R2, . . .,Rm

CPU cycles, memory space, I/O devices

Each resource typeRihas1 or moreinstances

Each process utilizes a resource as follows:

request use

release


8/43

7.2 Deadlock Characterization7.2 Deadlock Characterization



Deadlock CharacterizationDeadlock Characterization

Mutual exclusion:only one process at a time can use aresource

Hold and wait:a process holding at least one resource iswaiting to acquire additional resources held by otherprocesses

No preemption:a resource can be released only voluntarilyby the process holding it after that process has completed itstask

Circular wait:there exists a set {P0,P1, ,P0} of waitingprocesses such thatP0is waiting for a resource that is held

byP1,P1is waiting for a resource that is held byP2, ,Pn1is waiting for a resource that is held byPn, andPnis waiting for a resource that is held byP0

Deadlock can arise if four conditions hold simultaneously.



Resource-Allocation GraphResource-Allocation Graph

V is partitioned into two types:

P= {P1,P2, ,Pn}, the set consisting of all

the processes in the system

R= {R1,R2, ,Rm}, the set consisting of allresource types in the system

request edge directed edgeP1Rj

assignment edge directed edgeRjPi

A set of verticesVand a set of edgesE.



Resource-Allocation Graph (Cont.)Resource-Allocation Graph (Cont.)

Process

Resource Type with 4 instances

Pirequests instance ofRj

Piis holding an instance ofRj

Pi

Pi

Rj

Rj



Resource Allocation Graph With A DeadlockResource Allocation Graph With A Deadlock

Before P3requested an

instance of R2

After P3requested an

instance of R2



Graph With A Cycle But No DeadlockGraph With A Cycle But No Deadlock

Process P4may release its instance of resource type R2. That

resource can then be allocated to P3, thereby breaking the cycle.



Relationship of cycles to deadlocksRelationship of cycles to deadlocks

If a resource allocation graph contains no cyclesno deadlock

If a resource allocation graph contains a cycle andif only one instance exists perresource typedeadlock

If a resource allocation graph contains a cycle and and if several instancesexists per resource typepossibility of deadlock


15/43

7.3 Methods for Handling Deadlocks7.3 Methods for Handling Deadlocks



Methods for Handling DeadlocksMethods for Handling Deadlocks

Prevention

Ensure that the system willneverenter a deadlock state

Avoidance

Ensure that the system willneverenter an unsafe state

Detection

Allow the system to enter a deadlock state and then recover

Do Nothing

Ignore the problem and let the user or system administrator respond to the problem;

used by most operating systems, including Windows and UNIX


17/43

7.4 Deadlock Prevention7.4 Deadlock Prevention



Deadlock PreventionDeadlock Prevention

Mutual Exclusion The mutual-exclusion condition musthold for non-sharable resources

Hold and Wait we must guarantee that whenever a

process requests a resource, it does not hold any otherresources

Require a process to request and be allocated all its resourcesbefore it begins execution, or allow a process to requestresources only when the process has none

Result: Low resource utilization; starvation possible

To prevent deadlock, we can restrain the ways that a request can be made



Deadlock Prevention (Cont.)Deadlock Prevention (Cont.)

No Preemption

If a process that is holding some resources requests anotherresource that cannot be immediately allocated to it, then allresources currently being held are released

Preempted resources are added to the list of resources for whichthe process is waiting

A process will be restarted only when it can regain its oldresources, as well as the new ones that it is requesting

Circular Wait impose a total ordering of all resource types, andrequire that each process requests resources in an increasing order ofenumeration. For example:

F(tape drive) = 1 F(disk drive) = 5 F(printer) = 12


20/43

7.5 Deadlock Avoidance7.5 Deadlock Avoidance



Deadlock AvoidanceDeadlock Avoidance

Simplest and most useful model requires that each process declarethemaximum numberof resources of each type that it may need

The deadlock-avoidance algorithm dynamically examines the

resource-allocation state to ensure that there can never be acircular-wait condition

A resource-allocationstateis defined by the number of availableand allocated resources, and the maximum demands of theprocesses

Requires that the system has some additionala prioriinformationavailable.

a priori: formed or conceived beforehand



Safe StateSafe State

When a process requests an available resource, the system mustdecide if immediate allocation leaves the system in a safe state

A system is in a safe state only if there exists a safe sequence

A sequence of processes is a safe sequence for

the current allocation state if, for each Pi, the resource requests

that Pican still make, can be satisfied by currently available

resources plus resources held by allPj, withj



Safe State (continued)Safe State (continued)

If a system is in safe stateno deadlocks

If a system is in unsafe statepossibility of deadlock

Avoidanceensure that a system will never enter an

unsafe state


24/43


Safe, Unsafe , Deadlock StateSafe, Unsafe , Deadlock State


25/43


Avoidance algorithmsAvoidance algorithms

For a single instance of a resource type, use a resource-allocation graph

For multiple instances of a resource type, use the bankersalgorithm


26/43


Resource-Allocation Graph SchemeResource-Allocation Graph Scheme

Introduce a new kind of edge called a claim edge

Claim edgePi Rjindicates that processPjmayrequest resourceRj; which is represented by a dashed line

A claim edge converts to a request edge when a processrequestsa resource

A request edge converts to an assignment edge when theresource isallocatedto the process

When a resource isreleasedby a process, an assignmentedge reconverts to a claim edge

Resources must beclaimeda prioriin the system

R All i G h ihR All ti G h ith


27/43


Resource-Allocation Graph withResource-Allocation Graph with

Claim EdgesClaim Edges

Request

edge

Assignmentedge

Claim

edge

Claimedge


28/43


Unsafe State In Resource-Allocation GraphUnsafe State In Resource-Allocation Graph

Assignmentedge

Request

edge

AssignmentedgeClaim

edge


29/43


Resource-Allocation Graph AlgorithmResource-Allocation Graph Algorithm

Suppose that processPirequests a resourceRj

The request can be granted only if converting therequest edge to an assignment edge does not resultin the formation of a cycle in the resource allocation

graph


30/43


Bankers AlgorithmBankers Algorithm

Used when there existsmultipleinstances of a resource type

Each process musta prioriclaim maximum use

When a process requests a resource, it may have to wait

When a process gets all its resources, it must return them in a finite

amount of time


31/43


Data Structures for the Bankers AlgorithmData Structures for the Bankers Algorithm

Available:Vector of lengthm. If available [j] =k, there arekinstancesof resource typeRjavailable.

Max: n x mmatrix. IfMax[i,j] =k, then processPimay request at most

kinstances of resource typeRj.

Allocation: nxmmatrix. If Allocation[i,j] =kthenPiis currently

allocatedkinstances ofRj.

Need: nxmmatrix. IfNeed[i,j] =k, thenPimay needkmore instances

ofRjto complete its task.

Need[i,j]= Max[i,j] Allocation[i,j]

Letn= number of processes, andm= number of resources types.


32/43

7.6 Deadlock Detection7.6 Deadlock Detection


33/43


Deadlock DetectionDeadlock Detection

For deadlock detection, the system must provide

An algorithm that examines the state of the system to detect whether adeadlock has occurred

And an algorithm to recover from the deadlock

A detection-and-recovery scheme requires various kinds of overhead

Run-time costs of maintaining necessary information and executing thedetection algorithm

Potential losses inherent in recovering from a deadlock


34/43


Single Instance of Each Resource TypeSingle Instance of Each Resource Type

Requires the creation and maintenance of await-forgraph Consists of a variant of the resource-allocation graph

The graph is obtained byremovingthe resource nodes from aresource-allocation graph andcollapsingthe appropriateedges

Consequently; all nodes are processes

PiPjifPiis waiting forPj.

Periodically invoke an algorithm that searches for a cycle inthe graph

If there is a cycle, there exists a deadlock An algorithm to detect a cycle in a graph requires an order ofn2operations, wherenis the number of vertices in the graph

R All i G h dW if G hR All ti G h dW itf G h


35/43


Resource-Allocation Graph and Wait-for GraphResource-Allocation Graph and Wait-for Graph

Resource-Allocation Graph Corresponding wait-for graph


36/43


Multiple Instances of a Resource TypeMultiple Instances of a Resource Type

Available:A vector of lengthmindicates the number ofavailable resources of each type.

Allocation:Annxmmatrix defines the number of resourcesof each type currently allocated to each process.

Request:Annxmmatrix indicates the current request ofeach process. IfRequest[ij] =k, then processPiis requesting

kmore instances of resource type.Rj.

Required data structures:


37/43


Detection-Algorithm UsageDetection-Algorithm Usage

When, and how often, to invoke the detection algorithm depends on:

How often is a deadlock likely to occur?

How many processes will be affected by deadlock when it happens?

If the detection algorithm is invoked arbitrarily, there may bemanycyclesin the resource graph and so we would not be able to tellwhich oneof

the many deadlocked processes caused the deadlock

If the detection algorithm is invoked for every resource request, such anaction will incur a considerableoverheadin computation time

A less expensive alternative is to invoke the algorithm when CPUutilization dropsbelow 40%, for example

This is based on the observation that a deadlock eventually cripples systemthroughput and causes CPU utilization to drop


38/43

7.7 Recovery From Deadlock7.7 Recovery From Deadlock


39/43


Recovery from DeadlockRecovery from Deadlock

Two Approaches

Process termination

Resource preemption


40/43


Recovery from Deadlock:Recovery from Deadlock:Process TerminationProcess Termination

Abort all deadlocked processes

This approach will break the deadlock, but at great expense

Abort one process at a time until the deadlock cycle is eliminated

This approach incurs considerable overhead, since, after each process isaborted, a deadlock-detection algorithm must be re-invoked to determinewhether any processes are still deadlocked

Many factors may affect which process is chosen for termination

What is the priority of the process?

How long has the process run so far and how much longer will the processneed to run before completing its task?

How many and what type of resources has the process used? How many more resources does the process need in order to finish its task?

How many processes will need to be terminated?

Is the process interactive or batch?


41/43


Recovery from Deadlock:Recovery from Deadlock:Resource PreemptionResource Preemption

With this approach, we successively preempt some resources fromprocesses and give these resources to other processes until thedeadlock cycle is broken

When preemption is required to deal with deadlocks, then threeissues need to be addressed:

Selecting a victim Which resources and which processes are to bepreempted?

Rollback If we preempt a resource from a process, what should bedone with that process?

Starvation How do we ensure that starvation will not occur? That is,how can we guarantee that resources will not always be preemptedfrom the same process?


42/43


SummarySummary

Four necessary conditions must hold in the system for a deadlock

to occur

Mutual exclusion

Hold and wait

No preemption

Circular wait

Four principal methods for dealing with deadlocks

Use some protocol to (1)preventor (2)avoiddeadlocks, ensuring thatthe system will never enter a deadlock state

Allow the system to enter a deadlock state, (3)detectit,andthenrecover

Recover byprocess terminationorresource preemption

(4) Do nothing;ignore the problem altogether and pretend thatdeadlocks never occur in the system (used by Windows and Unix)

To prevent deadlocks, we can ensure thatat least oneof the fournecessary conditionsnever holds


43/43

End of Chapter 7End of Chapter 7

5. deadlocks

Documents