Transcript deadlock

Operating Systems
COMP 4850/CISG 5550
Deadlocks
Dr. James Money
Deadlocks
• Computer contain many resources
– Some can be shared
– The rest cannot be shared
• Examples of non-shared resources are
– Printers
– Tape drives
– Special memory addresses
– File access
Deadlocks
• Example: Printing
– Imagine if two processes had access to the
printer simultaneously
– One line might from process one
– The second might be from process two
– And repeating…
• Solution: grant temporary exclusive access
to a shared resource
Deadlocks
• Many times, processes need access to 2+
•
•
•
•
resources at the same time
Two processes are scanning a document and
writing it to CD
First process scans, then tries to write to CD
Second process locks the CD first, then tries to
scan
Both processes block! This is called a deadlock.
Deadlocks
• Same situation can occur across multiple
machines as well
• Many times on a network, these same
devices are shared and the same problem
occurs
• It may involve 3,4,5+ users and machines
Resources
• Deadlocks occur when processes exclusive
access to devices, file, etc.
• To simplify, we refer to any object that can
grant exclusive access as a resource.
• In a given computer system there are 10s100s resources to be managed.
Preemptable and Nonpreemptable
Resources
• There are two types of resources:
– Preemptable – resource that can be taken
away from a process without any ill effects
• Memory
• Usually does not cause deadlocks
– Nonpreemtable – resource that cannot be
taken away from a process without having a
computation fail
• CD Writer
• We focus on this for deadlocks
Resources
• The sequence to use a resource is:
– Request the resource
– Use the resource
– Release the resource
• If the resource is not available when requesting,
the process is forced to wait, either:
– Blocking
– Returns an error code and process tries again later
Resource Acquisition
• For some resources, such as a database
system, the user process can manage the
resources
• One way is to use a semaphore, as in
critical regions
Resource Acquisition
typedef int semaphore;
semaphore resource1;
void processA(void){
down(&resource1);
use_resource1();
up(&resource1);
}
Resource Acquisition – 2 Resources
typedef int semaphore;
semaphore resource1;
semaphore resource2;
void processA(void){
down(&resource1);
down(&resource2);
usebothresources();
up(&resource1);
up(&resource2);
}
Resource Acquisition
• Only one process involved so far
• What happens when we go to two
processes?
Resource Acquisition
typedef int semaphore;
semaphore resource1;
semaphore resource2;
void processA(void){
down(&resource1);
down(&resource2);
usebothresources();
up(&resource2);
up(&resource1);
}
void processB(void){
down(&resource1);
down(&resource2);
usebothresources();
up(&resource2);
up(&resource1);
}
Resource Acquisition
typedef int semaphore;
semaphore resource1;
semaphore resource2;
void processA(void){
down(&resource1);
down(&resource2);
usebothresources();
up(&resource2);
up(&resource1);
}
void processB(void){
down(&resource2);
down(&resource1);
usebothresources();
up(&resource1);
up(&resource2);
}
Introduction to Deadlocks
• Deadlocks are formally defined by
– A set of processes is deadlocked if each
process in the set is waiting for an event that
only another process in the set can cause
• Since they are all waiting, none of them
will wake up
• Assumption of no interrupts and single
threads
Conditions for Deadlock
1. Mutual Exclusion – each resource is either
2.
3.
4.
currently assigned to one process or is
available
Hold and Wait – processes currently holding
resources can request new resources
No preemption – Resources previously granted
cannot forcibly be taken away from the
process. They must be released by the process
Circular Wait – there must be a circular chain
of 2+ processes, each whom is waiting for a
resource held by the next member of the chain
Conditions for Deadlock
• All four conditions must exist for a
deadlock to occur
• If one is absent, deadlock cannot occur
• Many of these are related to system policy
choices
Deadlock Modeling
• Holt showed in 1972 that the four
conditions could be modeled using direct
graphs
• The graphs have two types of nodes:
– Processes – shown as circles
– Resources – shown as squares
Deadlock Modeling
• The arc from process to graph and vice
versa determines if the resource is held or
being waited on:
– An arc from a resource (square) to a process
(circle) indicates the resource is held by that
process
– An arc from a process to a resource indicates
the process is blocked waiting for the
resource
Deadlock Modeling
Dealing with Deadlocks
• Ignore the problem, maybe it will ignore
you?
– Used by UNIX and Windows
• Detection and Recovery
• Dynamic avoidance by careful resource
allocation
• Prevention by structurally negating one of
the four conditions for deadlocks
Ostrich Algorithm
• Here we stick our head in the sand like an
ostrich and pretend there is no problem
• Different reactions
– Mathematicians say it is unacceptable and
must be prevented all costs
– Engineers might ask how often does it really
occur and how serious the situation is
Ostrich Algorithm
• Most operating systems suffer from this
approach
• Example: Unix Process table
– Fixed length and finite
– If fork() fails because the table is full,
program waits and tries again (/bin/login does
this, for example)
Ostrich Algorithm
• However, consider this:
• Assume 100 process table slots
• Ten programs run that need to create 12
sub-processes each
• After each program creates 9 processes,
the table is full
• No process finishes because it keeps
trying to fork!
Ostrich Algorithm
• Do we abandon the processes and the
fork calls?
• Similar to this is the maximum number of
open files, which is restricted by the size
of the i-node table
• The solution is not obvious or easy, which
is why most operating systems choose this
approach