Download Course: Operating Systems Instructor: M Umair

Course: Operating Systems
Instructor: M Umair
M Umair – http://www.m-umair.com
Deadlock
M Umair – http://www.m-umair.com
Introduction
Under normal operation, a resource allocations proceed like this
1. Request a resource (suspend until available if necessary ).
2. Use the resource.
3. Release the resource.
{ Ref: http://web.cs.wpi.edu/~cs3013/c07/lectures/Section07-Deadlocks.pdf }
M Umair – http://www.m-umair.com
Introduction
A set of processes is in a deadlocked state when every process in
the set is waiting for an event that can be caused only by another
process in the set.
{ Ref: Operating System Concepts 8th Edition | Abraham Silberschatz , Greg Gagne , Peter B. Galvin }
M Umair – http://www.m-umair.com
The 4 necessary conditions
Mutual exclusion
One or more than one resource must be held by a process in a
non-sharable (exclusive) mode.
Hold and Wait
A process holds a resource while waiting for another resource
held by other process.
No Preemption
There is only voluntary release of a resource - nobody else can
make a process give up a resource.
Circular Wait
Process A waits for Process B waits for Process C .... waits for
Process A.
{ Ref: http://web.cs.wpi.edu/~cs3013/c07/lectures/Section07-Deadlocks.pdf }
M Umair – http://www.m-umair.com
Resource Allocation Graph
{ Ref: Operating System Concepts 8th Edition | Abraham Silberschatz , Greg Gagne , Peter B. Galvin }
M Umair – http://www.m-umair.com
Handling deadlock
We can deal with the deadlock problem in one of three ways:
• We can use a protocol to prevent or avoid deadlocks,
ensuring that the system will never enter a deadlocked state.
• We can allow the system to enter a deadlocked state, detect
it, and recover.
• We can ignore the problem altogether and pretend that
deadlocks never occur in the system.
{ Ref: Operating System Concepts 8th Edition | Abraham Silberschatz , Greg Gagne , Peter B. Galvin }
M Umair – http://www.m-umair.com
Deadlock Prevention
Mutual Exclusion
The mutual-exclusion condition must hold for non-sharable
resources. For example, a printer cannot be simultaneously shared
by several processes.
Sharable resources, in contrast, do not require mutually exclusive
access and thus cannot be involved in a deadlock. Read-only files
are a good example of a sharable resource.
We cannot prevent deadlocks by denying the mutual-exclusion
condition, because some resources are intrinsically nonsharable.
{ Ref: Operating System Concepts 8th Edition | Abraham Silberschatz , Greg Gagne , Peter B. Galvin }
M Umair – http://www.m-umair.com
Deadlock Prevention
Hold and Wait
To ensure that the hold-and-wait condition never occurs
One protocol that can be used requires each process to
request and be allocated all its resources before it begins
execution. Problems?
Solution: A process may request some resources and use
them. Before it can request any additional resources, however,
it must release all the resources that it is currently allocated.
{ Ref: Operating System Concepts 8th Edition | Abraham Silberschatz , Greg Gagne , Peter B. Galvin }
M Umair – http://www.m-umair.com
Deadlock Prevention
DVD Drive
Hard Drive
Printer
Problems?
>Device hold
>Starvation
DVD Drive
Hard Drive
Printer
{ Ref: Operating System Concepts 8th Edition | Abraham Silberschatz , Greg Gagne , Peter B. Galvin }
M Umair – http://www.m-umair.com
Deadlock Prevention
No Preemption
If a process is holding some resources and requests another
resource that cannot be immediately allocated to it (that is, the
process must wait), then all resources the process is currently
holding are preempted.
{ Ref: Operating System Concepts 8th Edition | Abraham Silberschatz , Greg Gagne , Peter B. Galvin }
M Umair – http://www.m-umair.com
Deadlock Prevention
Circular Wait
Let R = {R1, R2, ..., Rm} be the set of resource types
Define a one-to-one function F: R→N, where N is the set of
natural numbers.
A process can initially request any number of instances of a
resource type —say, Ri . After that, the process can request
instances of resource type Rj if and only if F(Rj ) > F(Ri ).
 Proof ?
{ Ref: Operating System Concepts 8th Edition | Abraham Silberschatz , Greg Gagne , Peter B. Galvin }
M Umair – http://www.m-umair.com
Deadlock Avoidance
Safe State
A state is safe if the system can allocate resources to each
process (up to its maximum) in some order and still avoid a
deadlock.
A safe state is not a deadlocked state. Conversely, a deadlocked
state is an unsafe state. Not all unsafe states are deadlocks.
An unsafe state may lead to a deadlock.
{ Ref: Operating System Concepts 8th Edition | Abraham Silberschatz , Greg Gagne , Peter B. Galvin }
M Umair – http://www.m-umair.com
Deadlock Avoidance
Total Resources: 12
Process
Maximum
Need
Currently
Held
Current
Need
Currently
Free
P0
10
5
5
-
P1
4
2
2
-
P2
9
2
5
-
Total
-
9
-
3
Process Sequence: P1, P0, P2 (Is it safe ?)
{ Ref: Operating System Concepts 8th Edition | Abraham Silberschatz , Greg Gagne , Peter B. Galvin }
M Umair – http://www.m-umair.com
Deadlock Avoidance
The Banker’s Algorithm
 A new process must declare the maximum number of instances
of each resource type that it may need.
 This number may not exceed the total number of resources in
the system
 System grants request only if the request will result into a safe
state.
{ Ref: http://www.cs.colostate.edu/~cs551/CourseNotes/Bankers.html}
M Umair – http://www.m-umair.com
Deadlock Avoidance
Banker Algorithm's Example - 1
A: 6k, B:5K, C:4k, D:7K deposits in bank. Bank maintain 10K cash in the branch
{ Ref: Modern Operating Systems, Andrew S. Tanenbaum }
M Umair – http://www.m-umair.com
Deadlock Avoidance
Banker Algorithm's Example -2
Assume we have the following resources
5 tape drives
2 graphic displays
4 printers
3 disks
A vector representing our total resources
Total = (5, 2, 4, 3)
{ Ref: http://www.cs.colostate.edu/~cs551/CourseNotes/Bankers.html}
M Umair – http://www.m-umair.com
Deadlock Avoidance
Allocation [4,2,2,3]
Total = (5, 2, 4, 3)
Allocation [4,2,2,3]
Available [1,0,2,0]
Need
{ Ref: http://www.cs.colostate.edu/~cs551/CourseNotes/Bankers.html}
M Umair – http://www.m-umair.com
Deadlock Recovery
Process Termination
Abort all deadlocked processes.
→Running for some time -> Resource/Time
→Results of partial computations -> Lost
→Re-compute later -> Resource/Time
The Problems
Abort one process at a time until the deadlock cycle is
eliminated.
→Considerable overhead
→What if process is updating a file ?
→Deadlock-detection algorithm must be invoked to check the
possibility of reoccurring deadlock.
{ Ref: Operating System Concepts 8th Edition | Abraham Silberschatz , Greg Gagne , Peter B. Galvin }
M Umair – http://www.m-umair.com
Deadlock Recovery
Considerations for process termination
1. Priority of the process
2. How long the process has computed and how much longer the process
will compute before completing its designated task
3. How many and what types of resources the process has used (whether
the resources are simple to preempt)
4. How many more resources the process needs in order to complete
5. How many processes will need to be terminated
6. Whether the process is interactive or batch
{ Ref: Operating System Concepts 8th Edition | Abraham Silberschatz , Greg Gagne , Peter B. Galvin }
M Umair – http://www.m-umair.com
Deadlock Recovery
Resource Preemption
 Selecting a victim
Determine the order of preemption to minimize cost
 Rollback




Safe state rollback
Restart it from that state
Difficult to determine what a safe state is
Keep information about the state of all running processes?
 Starvation
How to ensure that resources will not always be preempted from the
same process?
{ Ref: Operating System Concepts 8th Edition | Abraham Silberschatz , Greg Gagne , Peter B. Galvin }
M Umair – http://www.m-umair.com