Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Silberschatz, Galvin and Gagne ©2007 Deadlocks  (How to Detect Them and Avoid Them) A: After you! B: No, after you! A: After you! B: No, after you!.....  to infinity

7.2 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Roadmap: Deadlocks  The Deadlock Problem  System Model  Deadlock Characterization  Methods for Handling Deadlocks Deadlock Prevention Deadlock Avoidance Deadlock Detection

7.3 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Objectives  To develop a precise description of deadlocks, which prevent sets of concurrent processes from completing their tasks  To present a number of different methods for preventing or avoiding deadlocks in a computer system.

7.4 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 The Deadlock Problem  Example #1 System has 2 disk drives. P 1 and P 2 each hold one disk drive and each needs another one.  Example #2 semaphores A and B, initialized to 1 P 0 P 1 acquire (A)acquire(B) acquire (B)acquire(A)  Generalizing: A set of blocked processes each holding a resource and waiting to acquire a resource held by another process in the set.

7.5 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Bridge Crossing: Real-Life Icelandic Example  Traffic only in one direction at each moment in time.  Each section of a bridge can be viewed as a resource.  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. How can we reason about deadlocks? We need models!

7.6 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 System Model  Resource types R 1, R 2,..., R m CPU cycles, memory space, I/O devices etc.  Each resource type R i has W i instances.  Each process utilizes a resource as follows: request use release  What about the other things a process might do?

7.7 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Deadlock Characterization  Mutual exclusion: only one process at a time can use a resource.  Hold and wait: a process holding at least one resource is waiting to acquire additional resources held by other processes.  No preemption: a resource can be released only voluntarily by the process holding it, after that process has completed its task.  Circular wait: there exists a set {P 0, P 1, …, P n } of waiting processes such that P 0 is waiting for a resource that is held by P 1, P 1 is waiting for a resource that is held by P 2, …, P n–1 is waiting for a resource that is held by P n, and P 0 is waiting for a resource that is held by P 0. Deadlock can arise if four conditions hold simultaneously.

7.8 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Resource-Allocation Graph: Definition  V is partitioned into two types: P = {P 1, P 2, …, P n }, the set consisting of all the processes in the system. R = {R 1, R 2, …, R m }, the set consisting of all resource types in the system.  request edge – directed edge P 1  R j  assignment edge – directed edge R j  P i A set of vertices V and a set of edges E. The resource-allocation graph is therefore a bipartite directed graph. What would be the graph representing the bridge-crossing example?

7.9 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Resource-Allocation Graph (Pictorially)  Process  Resource type with 4 instances  P i requests instance of R j  P i is holding an instance of R j PiPi PiPi RjRj RjRj

7.10 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Example of a Resource Allocation Graph Is {P1,P2} a deadlock? Does the system have deadlocks? Deadlock: A set of blocked processes each holding a resource and waiting to acquire a resource held only by other processes in the set.

7.11 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Resource Allocation Graph With A Deadlock What is the “set of processes” witnessing the deadlock? Does the presence of a cycle in the graph imply a deadlock?

7.12 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Graph With A Cycle But No Deadlock Why is there no deadlock here?

7.13 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Basic Facts (Summary)  If graph contains no cycles  no deadlock. (Why?)  If graph contains a cycle  If only one instance per resource type, then deadlock. If several instances per resource type, possibility of deadlock.

7.14 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Methods for Handling Deadlocks  Ensure that the system will never enter a deadlock state.  Allow the system to enter a deadlock state, detect the situation, and then recover.  Ignore the problem and pretend that deadlocks never occur in the system; used by most operating systems, including UNIX.

7.15 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Deadlock Prevention  Mutual Exclusion – not required for sharable resources; must hold for nonsharable resources.  Hold and Wait – must guarantee that whenever a process requests a resource, it does not hold any other resources. Require process to request and be allocated all its resources before it begins execution, or allow process to request resources only when the process has none. Low resource utilization; starvation possible. Idea: Constrain the ways in which requests can be made.

7.16 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Deadlock Prevention (Cont.)  No Preemption – If a process that is holding some resources requests another resource that cannot be immediately allocated to it, then all resources currently being held are released. Preempted resources are added to the list of resources for which the process is waiting. Process will be restarted only when it can regain its old resources, as well as the new ones that it is requesting.  Circular Wait – impose a total ordering of all resource types, and require that each process requests resources in an increasing order of enumeration. Why increasing?

7.17 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Deadlock Avoidance  Simplest and most useful model requires that each process declare the maximum number of 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 a circular-wait condition.  Resource-allocation state is defined by the number of available and allocated resources, and the maximum demands of the processes. Requires that the system has some additional a priori information available.

7.18 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Safe State  When a process requests an available resource, system must decide if immediate allocation leaves the system in a safe state.  System is in safe state if there exists a sequence of ALL the processes in the system such that for each P i, the resources that P i can still request can be satisfied by currently available resources + resources held by all the P j, with j < i.  What does this say about P 1 ?  In what sense is this safe?  Is the system resulting from P 1 P 2 S.acquire(); Q.acquire(); Q.acquire(); S.acquire(); in a safe state?

7.19 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Basic Facts  If a system is in safe state  no deadlocks.  If a system is in unsafe state  possibility of deadlock.  Avoidance  ensure that a system will never enter an unsafe state. Deadlock-freedom is a safety property.

7.20 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Avoidance algorithms  Single instance of a resource type. Use a resource-allocation graph. Basic idea: Resources can be granted only if this does not create cycles in the resource-allocation graph. Why? Requires checking for cycles in directed graphs: quadratic algorithm in the number of processes.  Multiple instances of a resource type. Use the banker’s algorithm.

7.21 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Banker’s Algorithm (Dijkstra, THE OS)  Assumptions: Multiple instances of resources. Each process must a priori claim its maximum use for each resource. 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.

7.22 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Data Structures for the Banker’s Algorithm  Available: Vector of length m. If available [j] = k, there are k instances of resource type R j available.  Max: n x m matrix. If Max [i,j] = k, then process P i may request at most k instances of resource type R j.  Allocation: n x m matrix. If Allocation[i,j] = k then P i is currently allocated k instances of R j.  Need: n x m matrix. If Need[i,j] = k, then P i may need k more instances of R j to complete its task. Need [i,j] = Max[i,j] – Allocation [i,j]. Let n = number of processes, and m = number of resources types.

7.23 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Safety Algorithm: Are We Safe? 1.Let Work and Finish be vectors of length m and n, respectively. Initialize: Work = Available Finish [i] = false for i = 0, 1, …, n Find an i such that both: (a) Finish [i] == false (b) Need i  Work If no such i exists, go to step 4. 3.Work = Work + Allocation i Finish[i] = true go to step 2. 4.If Finish [i] == true for all i, then the system is in a safe state. Complexity?

7.24 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Resource-Request Algorithm for Process P i Request = request vector for process P i. If Request i [j] = k then process P i wants k instances of resource type R j. 1.If Request i  Need i go to step 2. Otherwise, raise error condition, since process has exceeded its maximum claim. 2.If Request i  Available, go to step 3. Otherwise P i must wait, since resources are not available. 3.Pretend to allocate requested resources to P i by modifying the state as follows: Available = Available – Request; Allocation i = Allocation i + Request i ; Need i = Need i – Request i ; l If safe  the resources are allocated to Pi. l If unsafe  Pi must wait, and the old resource- allocation state is restored

7.25 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Example of Banker’s Algorithm  5 processes P 0 through P 4 ; 3 resource types: A (10 instances), B (5 instances), and C (7 instances).  Snapshot at time T 0 : Allocation Max Available A B CA B C A B C P P P P P What is Need? Is the system in a safe state?

7.26 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Example: P 1 Requests (1,0,2) AllocationNeedAvailable A B CA B CA B C P P P P P  Executing safety algorithm shows that sequence satisfies safety requirement.  Can request for (3,3,0) by P 4 be granted?  Can request for (0,2,0) by P 0 be granted?

7.27 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Deadlock Detection  Allow system to enter deadlock state  Detection algorithm  Recovery scheme

7.28 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Single Instance of Each Resource Type  Maintain wait-for graph Nodes are processes. P i  P j if P i is waiting for P j.  Periodically invoke an algorithm that searches for a cycle in the graph. If there is a cycle, there exists a deadlock.  An algorithm to detect a cycle in a graph requires an order of n 2 operations, where n is the number of vertices in the graph.

7.29 Silberschatz, Galvin and Gagne ©2007 Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Resource-Allocation Graph and Wait-for Graph Resource-Allocation GraphCorresponding wait-for graph Why do we do it on the wait-for graph?

Operating System Concepts with Java – 7 th Edition, Nov 15, 2006 Silberschatz, Galvin and Gagne ©2007 End of OS in a Nutshell Hand-over Time Luca  Whoever follows Thanks for your cooperation during the lectures and good luck with the rest of the course!