Lecture 6: Concurrency: Mutual Exclusion and Synchronization

1
Lecture 6 Concurrency Mutual Exclusion and
Synchronization

• Operating System
• Fall 2006

2
Concurrency

• An OS has many concurrent processes that run in
  parallel but share common access
• Race Condition A situation where several
  processes access and manipulate the same data
  concurrently and the outcome of the execution
  depends on the particular order in which the
  access takes place.

3
Example for Race condition

• Suppose a customer wants to book a seat on UAL
  56. Ticket agent will check the -of-seats. If it
  is greater than 0, he will grab a seat and
  decrement -of-seats by 1.

4
Example for Race condition(cont.)
Ticket Agent 1 P1 LOAD -of-seats P2 DEC 1 P3
STORE -of-seats
Ticket Agent 2 Q1 LOAD -of-seats Q2 DEC 1 Q3
STORE -of-seats
Ticket Agent 3 R1 LOAD -of-seats R2 DEC 1 R3
STORE -of-seats
Suppose, initially, -of-seats12 Suppose
instructions are interleaved as
P1,Q1,R1,P2,Q2,R2,P3,Q3,R3 The result would be
-of-seats11, instead of 9
To solve the above problem, we must make sure
that P1,P2,P3 must be completely executed before
we execute Q1 or R1, or Q1,Q2,Q3 must be
completely executed before we execute P1 or R1,
or R1,R2,R3 must be completely executed before we
execute P1 or Q1.
5
Critical Section Problem
Critical section a segment of code in which the
process may be changing common variables,
updating a table, writing a file, and so on.
Goal To program the processes so that, at any
moment of time, at most one of the processes is
in its critical section.
6
Solution to Critical-Section Problem

• Any facility to provide support for mutual
  exclusion should meet the following requirements
• Mutual exclusion must be enforced Only one
  process at a time is allowed into its critical
  section
• A process that halts in its noncritical section
  must do so without interfering with other
  processes.
• A process waiting to enter its critical section
  cannot be delayed indefinitely
• When no process is in a critical section, any
  process that requests entry to its critical
  section must be permitted to enter without delay.
• No assumption are made about the relative process
  speeds or the number of processors.
• A process remains inside its critical section for
  a finite time only.

7
Three Environments

1. There is no central program to coordinate the
   processes. The processes communicate with each
   other through global variable.
2. Special hardware instructions
3. There is a central program to coordinate the
   processes.

8
Three Environments

1. There is no central program to coordinate the
   processes. The processes communicate with each
   other through global variable.
2. Special hardware instructions
3. There is a central program to coordinate the
   processes.

9
1st Attempt
Start with just 2 processes, P0 and p1 Global
variable turn, initially turn0
Prefix0 While (turn?0) do
CS0 turn1 suffix0
Prefix1 While (turn?1) do
CS1 turn0 suffix1
The processes take turn to enter its critical
section If turn0, P0 enters If turn1, P1 enters
This solution guarantees mutual exclusion.
But the drawback is that if one process leaves
the system or fails, the other will be blocked
permanently.
10
2st Attempt
Global variable flag0 and flag1, initially
flag0 and flag1 are both false
Prefix0 While (flag1) do flag0true
CS0 flag0false suffix0
Prefix1 While (flag0) do flag1true
CS1 flag1 false suffix1
If P0 is in critical section, flag0 is true
If P1 is in critical section, flag1 is true
If one process leaves the system, it will not
block the other process.
However, mutual exclusion is not guaranteed. P0
executes the while statement and finds that
flag1 is false P1 executes the while statement
and finds that flag0 is false. P0 sets flag0
to true and enters its critical section P1 sets
flag1 to true and enters its critical section.
11
3st Attempt
Global variable flag0 and flag1, initially
flag0 and flag1 are both false
Prefix0 flag0true While (flag1) do
CS0 flag0false suffix0
Prefix1 flag1true While (flag0) do
CS1 flag1 false suffix1
If P0 is in critical section, flag0 is true
If P1 is in critical section, flag1 is true
Guarantees mutual exclusion.
But mutual blocking can occur. P0 sets flag0 to
be true P1 sets flag1 to be true Both will be
hung in the while loop.
12
4st Attempt
Global variable flag0 and flag1, initially
flag0 and flag1 are both false
Prefix0 L0 flag0true If (flag1) then
flag0false goto L0
CS0 flag0false suffix0
Prefix1 L1 flag1true If (flag0) then
flag1false goto L1 CS1 flag1 false
suffix1
Guarantees mutual exclusion.
mutual blocking can occur if they execute at the
same speed.
13
Correct Solution
Initially, flag0flag1false turn0
Prefix0 flag0true while (flag1) do if
(turn1) flag0false while(turn1)
do flag0true CS0 turn1 flag0fals
e suffix0
Prefix1 flag1true while (flag0) do if
(turn0) flag1false while(turn0)
do flag1true CS1 turn0 flag1fals
e suffix1
14
Petersons Algorithm for 2 processes
Initially, flag0flag1false
Prefix0 flag0true turn1 while (flag1 and
turn1) do CS0 flag0false suffix0
Prefix1 flag1true turn0 while (flag0 and
turn0) do CS1 flag1false suffix1
15
Solution for n processes

• Global Variable
• Flag0..n-1 array of size n.
• Turn. Initially, Turnsome no. between 0 and n-1

Idle if Pi is outside Csi
Want-in if Pi wants to be in CSi
Flagi
in-CS if Pi is in CSi
16
Solutions for n processes
Pi Prefixi
Repeat Flagiwant-in jTurn while
j?i do if Flagj?idle then jTurn else j(j1)
mod n Flagiin-CS j0 while (jltn)
and (ji or Flagj?in-CS) do jj1 Until (j?n)
and (Turni or FlagTurnidle) Turni
CSi
j(Turn1)mod n While (j?Turn) and (Flagjidle)
doj(i1) mod n Turnj Flagiidle
17
Three Environments

1. There is no central program to coordinate the
   processes. The processes communicate with each
   other through global variable.
2. Special hardware instructions
3. There is a central program to coordinate the
   processes.

18
Hardware Support
Disable interrupt CS Enable interrupt
Wont work if we have multiprocessors
19
Special Machine Instructions

• Modern machines provide special atomic hardware
  instructions
• Atomic non-interruptable
• Either test memory word and set value
• Or swap contents of two memory words

20
TS Test and Set
Boolean TS(i) true if i0 it will also set i
to 1 false if i1
Initially, lock0
Pi Prefixi While( TS(lock)) do
CSi Lock0 suffixi
It is possible that a process may starve if 2
processes enter the critical section arbitrarily
often.
21
TS Test and Set (cont.)
To avoid the starvation problem, we can do the
following
Global variables waiting0..n-1 (boolean)
lock (integer) Local variable
Keyi (boolean)
Initially, lock0 and waitingifalse for all
0?i?n-1
Pi Waitingitrue Keyitrue While(waiting
i and Keyi) do Keyi TS(lock) Waitingifals
e CSi j(i1) mod n While(j?i and
waitingj) do j(j1) mod n if (ji) then
lock0 else Waitingjfalse
22
Exchange(int register, int memory)
Exchange the contents of a register with that of
memory.
Shared Variable lock, initially 0 local
variable key
Process Pi Prefixi Keyi1 While(Keyi?0)
do exchange(Keyi,lock) CSi Lock0
suffixi
23
Three Environments

1. There is no central program to coordinate the
   processes. The processes communicate with each
   other through global variable.
2. Special hardware instructions
3. There is a central program to coordinate the
   processes.

24
Semaphores

• A variable that has an integer value upon which 3
  operations are defined.
• Three operations
• A semaphore may be initialized to a nonnegative
  value
• The wait operation decrements the semaphore
  value. If the value becomes negative, then the
  process executing the wait is blocked
• The signal operation increments the semaphore
  value. If the value is not positive, then a
  process blocked by a wait operation is unblocked.
• Other than these 3 operations, there is no way to
  inspect or manipulate semaphores.

25
Wait(s) and Signal(s)

• Wait(s) is also called P(s)
• ss-1
• if (slt0) place this process in a waiting
  queue
• Signal(s) is also called V(s)
• ss1
• if(s?0) remove a process from the waiting
  queue

26
Semaphore as General Synchronization Tool

• Counting semaphore integer value can range over
  an unrestricted domain
• Binary semaphore integer value can range only
  between 0 and 1 can be simpler to implement
• Also known as mutex locks
• Wait B(s) s is a binary semaphore
• if s1 then s0 else block this process
• Signal B(s)
• if there is a blocked process then unblock
  a process else s1
• Can implement a counting semaphore S as a binary
  semaphore

27
Note

• The wait and signal primitives are assumed to be
  atomic they cannot be interrupted and each
  routine can be treated as an indivisible step.

28
Mutual Exclusion provided by Semaphores

• Semaphore S // initialized to 1
• Pi
• prefixi
• wait (S)
• CSi
• signal (S)
• suffixi

• Signal B(s)
• if there is a blocked process
• then unblock a process
• else s1

• Wait B(s) s is a binary semaphore
• if s1
• then s0
• else block this process

29
Message Passing

• Direct Addressing
• Specific identifier of source and destination
  processes
• Send(destination, message)
• Receive(source, message)
• Indirect Addressing
• Messages are not sent directly from sender to
  receiver but rather are sent to a shared data
  structure consisting of queues that can
  temporarily hold messages.
• Such queues are generally referred to as
  mailboxes.
• Thus, for 2 processes to communicate, one process
  sends a message to the appropriate mailbox and
  the other process picks up the message from the
  mailbox.

30
Message Passing (cont.)

• When a send primitive is executed in a process,
  there are 2 possibilities
• Either the sending process is blocked until the
  message is received
• Or it is not
• When a receive primitive is executed in a
  process, there are 2 possibilities
• If a message has previously been sent, the
  message is received and execution continues
• If there is no waiting message, then either
• The process is blocked until a message arrives,
  or
• The process continues to execute, abandoning the
  attempt to receive.
• Blocking send, blocking receive
• Nonblocking send, blocking receive
• Nonblocking send , nonblocking receive

31
Mutual Exclusion

• Create_mailbox(mutex)
• Send(mutex, null)
• Pi
• Prefixi
• Receive(mutex,msg)
• CSi
• Send(mutex,msg)
• Suffixi

Main process Mailbox name is mutex
32
Two classical examples

• Producer and Consumer Problem
• Readers/Writers Problem

33
Two classical examples

• Producer and Consumer Problem
• Readers/Writers Problem

34
Producer and Consumer Problem

• Producer can only put something in when there is
  an empty buffer
• Consumer can only take something out when there
  is a full buffer
• Producer and consumer are concurrent processes

0








consumer
producer
N buffers
N-1
35
Producer and Consumer Problem(cont.)

• Global Variable
• B0..N-1 an array of size N (Buffer)
• P a semaphore, initialized to N
• C a semaphore, initialized to 0
• Local Variable
• In a ptr(integer) used by the producer, in0
  initially
• Out a ptr(integer) used by the consumer, out0
  initially

36
Two classical examples

• Producer and Consumer Problem
• Readers/Writers Problem

37
Readers/Writers Problem

• Suppose a data object is to be shared among
  several concurrent processes. Some of these
  processes want only to read the data object,
  while others want to update (both read and write)
• Readers Processes that read only
• Writers processes that read and write
• If a reader process is using the data object,
  then other reader processes are allowed to use it
  at the same time.
• If a writer process is using the data object,
  then no other process (reader or writer) is
  allowed to use it simultaneously.

38
Solve Readers/Writers Problem using wait and
signal primitives(cont.)

• Global Variable
• Wrt is a binary semaphore, initialized to 1 Wrt
  is used by both readers and writers
• For Reader Processes
• Mutex is a binary semaphore, initialized to
  1Readcount is an integer variable, initialized
  to 0
• Mutex and readcount used by readers only

39
End of lecture 6
Thank you!