Chapter 4 Processes

 <back>


         Process Concept

         Process Scheduling

         Operation on Processes

         Cooperating Processes

         Interprocess Communication

 


Process Concept

 

         An operating system executes a variety of programs:

        Batch system – jobs

        Time-shared systems – user programs or tasks

         Textbook uses the terms job and process almost interchangeably.

         Process – a program in execution; process execution must progress in sequential fashion.

         A process includes:

        program counter

        stack

        data section

         As a process executes, it changes state

        new:  The process is being created.

        running:  Instructions are being executed.

        waiting:  The process is waiting for some event to occur.

        ready:  The process is waiting to be assigned to a process.

        terminated:  The process has finished execution.

Information associated with each process.

         Process state

         Program counter

         CPU registers

         CPU scheduling information

         Memory-management information

         Accounting information

         I/O status information

 


         Process Scheduling

 

         Job queue – set of all processes in the system.

         Ready queue – set of all processes residing in main memory,
ready and waiting to execute.

         Device queues – set of processes waiting for an I/O device.

         Process migration between the various queues.

         Long-term scheduler (or job scheduler) – selects which processes should be brought into the ready queue.

         Short-term scheduler (or CPU scheduler) – selects which process should be executed next and allocates CPU.

         Short-term scheduler is invoked very frequently (milliseconds)
Þ (must be fast).

         Long-term scheduler is invoked very infrequently (seconds, minutes) Þ (may be slow).

         The long-term scheduler controls the degree of multiprogramming.

         Processes can be described as either:

         I/O-bound process – spends more time doing I/O than computations, many short CPU bursts.

         CPU-bound process – spends more time doing computations; few very long CPU bursts.

         When CPU switches to another process, the system must save the state of the old process and load the saved state for the new process.

         Context-switch time is overhead; the system does no useful work while switching.

         Time dependent on hardware support.

  


         Operation on Processes

          

         Parent process creates children processes, which, in turn create other processes, forming a tree of processes.

         Resource sharing

         Parent and children share all resources.

         Children share subset of parent’s resources.

         Parent and child share no resources.

         Execution

         Parent and children execute concurrently.

         Parent waits until children terminate.

         Address space

         Child duplicate of parent.

         Child has a program loaded into it.

         UNIX examples

         fork system call creates new process

         execve system call used after a fork to replace the process’ memory space with a new program.

         Process executes last statement and asks the operating system to decide it (exit).

         Output data from child to parent (via wait).

         Process’ resources are deallocated by operating system.

         Parent may terminate execution of children processes (abort).

         Child has exceeded allocated resources.

         Task assigned to child is no longer required.

         Parent is exiting.

         Operating system does not allow child to continue if its parent terminates.

         Cascading termination.

 


         Cooperating Processes

 

         Independent process cannot affect or be affected by the execution of another process.

         Cooperating process can affect or be affected by the execution of another process

         Advantages of process cooperation

        Information sharing

        Computation speed-up

        Modularity

        Convenience

         Paradigm for cooperating processes, producer process produces information that is consumed by a consumer process.

        unbounded-buffer places no practical limit on the size of the buffer.

        bounded-buffer assumes that there is a fixed buffer size.

         Shared data

 

var n;

type item = … ;

var buffer. array [0..n–1] of item;

in, out: 0..n–1;

         Producer process

repeat

produce an item in nextp

while in+1 mod n = out do no-op;

buffer [in] :=nextp;

in :=in+1 mod n;

until false;

         Consumer process

repeat

while in = out do no-op;

nextc := buffer [out];

out := out+1 mod n;

       

consume the item in nextc

         

until false;

 

         Solution is correct, but can only fill up n–1 buffer.

 

         A thread (or lightweight process) is a basic unit of CPU utilization; it consists of:

        program counter

        register set

        stack space

         A thread shares with its peer threads its:

        code section

        data section

        operating-system resources

collectively know as a task.

         A traditional or heavyweight process is equal to a task with one thread

         In a multiple threaded task, while one server thread is blocked and waiting, a second thread in the same task can run.

         Cooperation of multiple threads in same job confers higher throughput and improved performance.

         Applications that require sharing a common buffer (i.e., producer-consumer) benefit from thread utilization.

         Threads provide a mechanism that allows sequential processes to make blocking system calls while also achieving parallelism.

         Kernel-supported threads (Mach and OS/2).

         User-level threads; supported above the kernel, via a set of library calls at the user level (Project Andrew from CMU).

         Hybrid approach implements both user-level and kernel-supported threads (Solaris 2).

 


         Interprocess Communication

 

         Mechanism for processes to communicate and to synchronize their actions.

         Message system – processes communicate with each other without resorting to shared variables.

         IPC facility provides two operations:

        send(message) – message size fixed or variable

        receive(message)

         If P and Q wish to communicate, they need to:

        establish a communication link between them

        exchange messages via send/receive

         Implementation of communication link

        physical (e.g., shared memory, hardware bus)

        logical (e.g., logical properties)

         How are links established?

         Can a link be associated with more than two processes?

         How many links can there be between every pair of communicating processes?

         What is the capacity of a link?

         Is the size of a message that the link can accommodate fixed or variable?

         Is a link unidirectional or bi-directional?

         Processes must name each other explicitly:

        send (P, message) – send a message to process P

        receive(Q, message) – receive a message from process Q

         Properties of communication link

        Links are established automatically.

        A link is associated with exactly one pair of communicating processes.

        Between each pair there exists exactly one link.

        The link may be unidirectional, but is usually bi-directional.

         Messages are directed and received from mailboxes (also referred to as ports).

        Each mailbox has a unique id.

        Processes can communicate only if they share a mailbox.

         Properties of communication link

        Link established only if processes share a common mailbox

        A link may be associated with many processes.

        Each pair of processes may share several communication links.

        Link may be unidirectional or bi-directional.

         Operations

        create a new mailbox

        send and receive messages through mailbox

        destroy a mailbox

         Mailbox sharing

        P1, P2, and P3 share mailbox A.

        P1, sends; P2 and P3 receive.

        Who gets the message?

         Solutions

        Allow a link to be associated with at most two processes.

        Allow only one process at a time to execute a receive operation.

        Allow the system to select arbitrarily the receiver.  Sender is notified who the receiver was.

         Queue of messages attached to the link; implemented in one of three ways.

1.     Zero capacity – 0 messages
Sender must wait for receiver (rendezvous).

2.     Bounded capacity – finite length of n messages
Sender must wait if link full.

3.     Unbounded capacity – infinite length
Sender never waits.