Outlook. Deadlock Characterization Deadlock Prevention Deadlock Avoidance

Transcription

1 Deadlocks

2 Outlook Deadlock Characterization Deadlock Prevention Deadlock Avoidance Deadlock Detection and Recovery e 2

3 Deadlock Characterization 3

4 Motivation System owns many resources of the types Memory, CPU, Files, and I/O Many resources require exclusive access; e.g. printer or CD-RW Usage pattern request R use R release R Allocation of more than one resource may lead to a deadlock 4

5 Deadlock Example Assume two processes P 1 and P 2 with P i starting to copy data from resource CD i to CD i+1 P1 Request CD1 P2 Request CD2 Request CD2 Request CD1 <Copy> <Copy> Release CD2 Release CD1 Release CD1 Release CD2 5

6 Deadlock Example Both may wait on each other to release the resource P1 Request CD1 Request CD2 P2 Request CD2 Request CD1 <Copy> <Copy> Release CD2 Release CD1 Release CD1 Release CD2 6

7 Necessary Conditions for a Deadlock Mutual exclusion only one process can use a resource at a time Hold and wait Process holds at least one resource and is waiting to acquire other ones No preemption p Process releases resources only after completing its task Circular wait There is a process ordering such that process P i waits for a resource allocated by P i+1 mod n 7

8 Resource Allocation Graph Vertices {P 1,,P n } the set of processes {R 1,,R m } the set of resources; each resource may have several instances; each instance is denoted by a dot P i R i Edges P i R k :P i requests an instance of R k R k P j : P j holds an instance of R k P i P j Rk 8

9 Resource Allocation Graph: Example 9

10 Resource Allocation Graph: Properties Deadlock Cycle 10

11 Resource Allocation Graph: Properties All resources have a single instance: Deadlock Cycle Cycle involves only single instance resources Deadlock (i.e. in such situation a cycle is a necessary and sufficient condition for a deadlock) 11

12 Cycle Deadlock? Deadlock No deadlock 12

13 Methods for handling deadlocks Use a protocol to prevent or avoid deadlocks, assuring that the system never enters a deadlock state Deadlock prevention provides a set of method to assure that at least one of the four necessary deadlock conditions cannot hold Deadlock avoidance decide for each process if it can proceed or has to wait based on available resources, currently allocated resources and the resources allocated by a process in future (processes have to know this in advance!) Allow the system to enter a deadlock state, detect it and recover Ignore the problem and pretend that deadlocks never occur in the system 13

14 Deadlock Prevention 14

15 Deadlock Prevention Idea: Ensure that one of the four necessary conditions must not hold Mutual exclusion Reasonable for sharable resources like read only files (However, other resources like a printer are intrinsically non-sharable and thus need mutual exclusion) Hold and Wait Request all resources prior to execution or Release all resources before allocating new ones 15

16 Deadlock Prevention No preemption Preempt all resources of a process requesting a resource which is already allocated Process is inserted in the waiting queue of preempted and the new requested resource Process can continue if all these resources are granted Alternative: preempt only these resources which are being requested during the wait Process is placed in waiting queue of requested resource Allocated resources are preempted p only if needed by another process 16

17 Deadlock Prevention Circular Wait Total ordering on resources F: {R 1, R m } IN Lock resources in ascending/descending order Alternative: all resources greater/smaller or equal have to be released before locking the new one Locking all resources in ascending order in fact prevents circular wait. Assume for the sake of contradiction a process set {P1,,Pn} such that Pi holds resource Ri and waits for resource Ri+1 which is held by Pi+1 (modulo n). Since Pi request resource Ri after Ri+1 (modulo n), we have: Thus: 17

18 Deadlock Prevention Circular Wait Ordering alone does not prevent circular wait It s up to the application programmers to follow the ordering System may issue a warning if ordering is not followed How to decide on the right ordering? Idea: ordering is determined dynamically while programs are executing Example: Lock-order verifier witness on BSD UNIX used in conjunction with mutual-exclusion locks 18

19 Deadlock Prevention (witness example) Thread 1: pthread_mutex_lock(&locka); pthread_mutex_lock(&lockb); /* do some work */ pthread_mutex_unlock(&lockb); t l kb) pthread_mutex_unlock(&locka); Thread 2: pthread_mutex_lock(&lockb); pthread_ mutex_ lock(&locka); /* do some work */ pthread_mutex_unlock(&locka); pthread_mutex_unlock(&lockb); unlock(&lockb); Lock ordering if Thread 1 executed first: Once Thread 2 executes, witness generates a warning on the console. 19

20 Deadlock Avoidance 20

21 Deadlock Avoidance: Safe States Assumption: each process declares the max. number of resources instances it needs Definition: P 1,,P n is a safe sequence if possible resource requests of P i can be satisfied by Currently available resources and Resources held by all P j for j<i System is in a safe state if one such sequence exists System is in an unsafe state if no such sequence exists deadlock unsafe System in safe state no deadlock can occur Deadlock system is in unsafe state System in unsafe state t not necessarily implies a deadlock safe 21

22 Example Consider 12 resource instances and three processes P 0, P 1, P 2 with the following allocations and maximum needs Maximum P P P Allocated 22

23 Example P 1 1, P 0 0, P 2 is a safe sequence Maximum Allocated Available P P 1 finishes P P P 0 finishes 23

24 A System can go from Safe to Unsafe State Example: assume P2 is granted an additional resource Maximum P P P Allocated Only 2 resources available only P1 can be allocated all resources 24

25 A System can go from Safe to Unsafe State Maximum Allocated Available P P P 1 finishes P P 0 may request 5 more resources P 0 has to wait P 2 may request 6 more resources P 2 has to wait 2 y q 2 Deadlock! 25

26 Deadlock Avoidance: General Idea System starts in a safe state For each request Grant resource if system remains in a safe state Process has to wait, otherwise We consider two example algorithms which assure the system staying in safe state Resource-allocation-graph algorithm Banker s algorithm 26

27 Resource-Allocation-Graph Algorithm Assumption: each resource has exactly one instance Idea: extend resource allocation Graph by claim edges Transitions: Claim Edge request Request Edge R R R released granted Assignment Edge P P P Grant a resource (either as assigned if it is free or as requested if it is held by another process) if the graph remains cycle free Otherwise process has to wait 27

28 Resource-Allocation-Graph Algorithm: Example R 2 may not be granted to P 2 (a) (b) 28

29 Banker s Algorithm Resource-allocation-graph algorithm not applicable when resources can have more than one instance. The following algorithm is applicable then as well. A process must declare in advance its maximum need for each resource type. This number may not exceed the number of available resources of fthese types. When a process request resources the system checks if granting the resources leaves the system in safe state If it will, resources are allocated Otherwise, the process must wait until other processes have released enough resources Before we proceed: we need the following data structures 29

30 Banker s Algorithm Let n be the number of processes and m be the number of resource types We declare the following vector of length m and the following matrices of size n x m Available[j]=k k instances of resource Rj available Max[i][j]=k Pi requests at most k instances of Rj Allocation[i][j]=k Pi is allocated k instances of Rj Need[i][j]=Max[i][j] - Allocation[i][j] 30

31 Banker s Algorithm Some notation before we proceed Let x = (1 (x1, xn) and y = (1 (y1,,yn) be two vectors of length n. We say x y iff x1 y1,, xn yn We say x < y if x y and x y Example: if x=(0,3,2,1) and y=(1,7,3,2) then x y and x < y. We treat each row in the matrices Allocation and Need as vectors of length m and refer to them as Allocation i and Need i Allocation i specifies the ressources currently allocated to process Pi Next i specifies the additional ressources that Pi may still request to complete ist task. 31

32 Banker s Algorithm: Safety Algorithm 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. 32

33 Banker s Algorithm: Request Algorithm Input: Resource request Request i by process P i, i.e., if Request i [j] == k then process P i wants k instances of resource type R j. p i yp j If Request i Need i we can continue, otherwise we are already done (error: the process has exceeded its maximum claim) If Request i Available we can continue, otherwise P i must wait, since the resources are not available Pretend resource allocation Available = Available Request i Allocation i = Allocation i + Request i Need i = Need i Request i Check for safe state (Safety Algorithm) P i gets resource if result state is safe Otherwise P i must wait Restore old state 33

34 Banker s Algorithm: Example Consider a system with fife processes P 0, P 1, P 2, P 3, P 4 and three resource types A, B, C. Instances A B C Available A B C Allocation Max Need A B C A B C P P P P P 0 A B C P P P P P

35 Banker s Algorithm: Example Instances A B C Available A B C Is the system in safe state? Allocation Max Fin A B C A B C ish P P P P P Need A B C P P P P P Consider the sequence: 35

36 Banker s Algorithm: Example Instances A B C Available A B C Consider now that process P1 requests one additional resource of type A and dtwo of ftype B. Allocation Max Fin Need A B C A B C ish A B C P P P P P P P P P P Request(1) = (,, ). First check that Request(1) Available. 36

37 Banker s Algorithm: Example Instances A B C Available A B C Consider now that process P1 requests one additional resource of type A and dtwo of ftype B. Allocation Max Fin Need A B C A B C ish A B C P P P P P P 1 P P P P Request(1) = (1,0,2).OldAllocation(1)=, (2,0,0), 0 Now pretend that the request has been fulfilled. 37

38 Banker s Algorithm: Example Instances A B C Available A B C Consider now that process P1 requests one additional resource of type A and dtwo of ftype B. Allocation Max Fin Need A B C A B C ish A B C P P P P P P P P P P Check if a safe sequence can be found. As done before we can find for instance <P1, P3, P4, P0, P2>. Exercise. 38

39 Banker s Algorithm: Example Instances A B C Available A B C Consider now that new system state. Can the following request be granted? Allocation Max Fin Need A B C A B C ish A B C P P P P P P P P P P Request(4) = ( 3, 3, 0). First check that Request(4) Available. 39

40 Banker s Algorithm: Example Instances A B C Available A B C Consider now that new system state. Can the following request be granted? Allocation Max Fin Need A B C A B C ish A B C P P P P P P P P P P Request(0)=( ( 0,2,0).ObviouslyRequest(0), 0 Available. But remains the system in a safe state? 40

41 Deadlock Detection and Recovery 41

42 Deadlock Detection System without deadlock prevention or avoidance deadlock may occur If deadlocks are to be handled, however, e the system must provide then Algorithm which examines whether a deadlock has occurred Algorithm to recover from a deadlock We consider two methods again Wait-for-graph: each resource has only one instance and Algorithm similar to bankers-algorithm when resources can have several instances 42

43 DD: Single Instance of a Resource Remove resource nodes Resource allocation graph Wait-for graph Deadlock wait for graph contains a circle Maintain wait-for-graph Periodically check for deadlock O(n^2) 43

44 DD: Several Instances of a Resource Type We consider m resource types and n processes. We define similar data structures as in the Banker s algorithm Available: A vector of length m indicates the number of available resources of each type. Allocation: An n x m matrix defines the number of resources of each type currently allocated to each process. Request: An n x m matrix indicates the current request of each process. If Request [i][j]= k, k then process P i is requesting k more instances of resource type R j. In the following we consider again the relation and dfor each processor Pi the rows of Allocation and Request as vectors Allocation i and Request i 44

45 DD: Several Instances of a Resource Type 1. Let Work and Finish be vectors of length m and n, respectively Initialize: (a) Work = Available (b) For i = 1,2,, n, if Allocation i 0, then Finish[i] [] = false;otherwise, Finish[i] [] = true. 2. Find an index i such that both: (a) Finish[i] == false (b) Request i Work If no such i exists, go to step 4. 45

46 DD: Several Instances of a Resource Type 3. Work = Work + Allocation i Finish[i] = true go to step If Finish[i] i == false, for some i, i 1 i n, then the system is in deadlock state. Moreover, if Finish[i] i == false, then P i is deadlocked. dl d 46

47 Example Instances A B C Available A B C Is the system in a deadlocked state? Allocation Request Fin A B C A B C ish P P P P P Consider the sequence: 47

48 Example Instances A B C Available A B C Suppose now that process P2 makes one additional request for an instance of type C. Allocation Request Fin A B C A B C ish P P P P P Is the system in a deadlocked state? We can reclaim the resources of Process P0. But then? 48

49 Detection-Algorithm Usage When to invoke the deadlock detection algorithm? In the extreme, whenever a requested resource could not be granted immediately. Advantage: in this case, the process causing the deadlock can be identified. Disadvantage: incurs a considerable overhead in computation time. Less expensive alternative: invoke the algorithm in less frequent intervals e.g. once per hour (deadlocks occur infrequently) or whenever CPU utilization drops below 40% (deadlock eventually cripples system throughput and causes CPU utilization drop) Once a deadlock was detected: how to recover from a deadlock? (Inform the user/operator: he can for instance manually terminate and restart processes or the whole system) System recovers automatically by process termination System recovers automatically by resource preemption 49

50 Recovery from Deadlock: Process Termination System reclaims all resources allocated by the terminated thread Two possible methods Abort all deadlocked processes Abort one process at a time What is a good sequence? Many Factors: Priority, remaining computation time, number and type of allocated/required resources, interactive/batch 50

51 Recovery from Deadlock: Resource Preemption Successively preempt resources and pass them to others Continue this until deadlock is broken Issues to be addressed Select victim which resources and which process are to be preempted? (consider for instance the time a process already computed and/or the time it still predicts its outstanding computation.) Rollback Process has to be rolled back to a safe state from where it can resume execution once it got all required resources (back) Starvation how to assure that resources are not always preempted from one process? (e.g. include the number of rollback in the cost factor when determining the process from which resources are preempted) 51

52 Summary and References 52

53 Summary Deadlock: two or more processes waiting indefinitely for an event which can only be caused by the waiting processes themselves Three principal methods to deal with deadlocks Protocols to avoid or prevent deadlocks, assuring that the system never enters a deadlock state Allow the system running into a deadlock situation, detect it at some point in time and recover Just ignore the problem as done in Windows or UNIX (deadlock occur infrequently) Four necessary and sufficient conditions for a deadlock: mutual exclusion, hold and wait, no preemption, and circular wait Idea of deadlock prevention: just assure that not all four condition can occur at the same time Deadlock avoidance algorithms: less stringent but require a priory information on how (many) resources might be used by a process. Deadlock detection alternatives: how often to invoke? what to do if a deadlock is detected? Whenever preemption is used, three issues must be addressed: selecting a victim, rollback and starvation. 53

54 References Silberschatz, Galvin, Gagne, Operating System Concepts, Seventh Edition, Wiley, 2005 Chapter 7 Deadlocks 54