# Outline:
1. process concept
2. process scheduling
3. operations on processes
4. Inter-process communication

1. process concept:

process = a program in execution

Process state:

1. new: The process is being created
2. running: Instructions are being executed
3. waiting: The process is waiting for some event to occur (I/O request)
4. ready : The process is waiting to be assigned to a processor
5. terminated : The process

PCB(process control block)

• 프로세스의 모든 정보를 담고 있는 자료구조( process state, program counter, cpu registers, cpu scheduling information, memory-management information , accounting information , i/0 status information 등등)

• context switch가 발생하면:

• 다른 프로세스로 전환될때 실행중이던 process의 모든 정보인 snapshot를 PCB에 저장해놓는다.
• 나중에 이 프로세스의 상태가 ready->running 으로 바뀌어도 PCB의 값을 보고 다시 실행을 이어나갈수있다.

2. Process scheduling:

• Scheduling queue: Process들을 담을 공간
  1. Job queue: 모든 프로세스들의 집합
  2. Ready queue: State가 Ready인 프로세스들의 집합
  3. Device queue: I/O 디바이스를 기다리는 프로세스의 집합
  프로세스들은 다양한 큐 사이를 이동한다.
• Scheduler:
  1. Long-term scheduler(Job scheduler):
     메모리를 어떻게 할당할지를 결정
     어떤 프로세스를 Ready queue에 넣을지 결정
  2. Short-term scheduler(CPU scheduler):
     빠르게 context switching을 해준다.
     Ready queue에서 프로세스를 선택
     concurrently처럼 보이게 해준다.
  3. Medium-term scheduler:
     Swapping을 담당한다.

Context switch:

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 a big overhead;

3. Operation on processes

Process creation(fork)
Process termination(exit)

Operation on process

• Resoource sharing options

  child share subset of parent's resources

• Execution options

  Parent and child execute concurrently

  Parent waits until child terminates (wait() 함수를 이용하여)

• Address space

  child duplicates parent's address space

  child has a program loaded into it(exec() 함수를 이용하여)

  

  

Fork system call

1. resource sharing:
   • parent and children share files
   • CPU time or memory(address space) are not shared (부모와자식프로세스는 별도의 프로세스이므로)
2. Execution
   • parent and children execute concurrently
3. Address space
   • Child Duplicate parent address space
4. Call once, return twice:
   • fork() return value의 값이 parent(=pid of child)와child(=0)에 따라 다르다.

• Every process has the same program!
  exec system call is used to load a new program
  
• Reaping child processes(부모가 자식을 거둔다)
  parent waits for its children to terminate
  

zombie process

자식 프로세스가 exit()하였는데 부모 프로세스가 wait() 하지않아 자식 프로세스가 zombie 상태가 되는것.

Process termination

• exit: 마지막문장에서 실행되고 OS에게 자발적으로 삭제해달라고 요청한다.

• abort: 부모가 자식 프로세스를 종료시키는 것.

위 그림에서 1번은 orpahn process가 발생한것이고, 2번은 cascading termination이 발생한것이다.

4. Inter-process communication(IPC)

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

• Cooperationg process can affect or be affected by the execution of another process.-> Process 간 communication이 필요하다.

• 장점:

  1. information sharing(ex:file)
  2. computation speed-up
  3. modularity
  4. convinience

IPC:

1. Message passing
2. Shared Memory

위 두가지 방법은 os의 도움이 있어야 한다.

• Producer-Consumer problem using a programmed shared-memory
• producer process produces information
• consumer consumes informathion
• Must be synchronized

Spinlock:

버퍼가 비어있거나, 가득차있을때 발생하는 것.
while(); 루프를 계속돌고 있는 상황

Shared memory:

1. shmget: 프로세스가 share memory를 만든다.

2. shmat: 프로세스가 자신의 address space를 shared memory에 붙인다.

Message-passing:

• IPC facility provides two operations:
  - send(message)
  - receive(message)
• If P and Q communicate:
  - establish Communication link (maibox) between them
  - exchange messages via send/receive
  

Establishing links:

1. Direct communication

   • process must name each other explicitly (= "hard-coding") -> 이름이 바뀌면 recompile 해야한다.
   • send(P,message) - send a message to process P
   • receive(Q,message) - receive a message from process Q

   

2. Indirect communication:

   • Messages are directed and received from mailboxes
   • each mailbox has a unique id
   • processes can communicate only if thy share a mailbox

   

Direct communication vs Indirect communication:

• Direct communication(pipe):
  - Links are established automatically
  - usually bi-directional
  - Between each pair there exists exactly one link
  - shortcoming: can we know the name in advance?

• Indirect communication
  - 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

1. A link may be associated with many processes.

2. Each pair of processes may share several commuication links.

Indirect communication operations:

1. create a new mailbox.
2. send and receive messages through mailbox.
3. destroy a mailbox.

who gets the message (if p2,p3 tries)?

1. allow a link to be assocaited with at most two processes.
2. Broadcasting
3. allow only one process to receive a message.
4. Allow the sending process to select the receiver.

Mailbox synchronization:

• Message passing may be either blocking or non-blocking.

• Blocking is considered Synchronous.

• Non-blocking is considered Asynchronous.

OS는 Primitive만 제공해주고, 옵션은 프로그래머가 선택한다.

IPC comparsion

Message VS Shared memory:

1. Message-passing:
   useful for inter-computer communication , exchanging smaller amounts of data and implemented using system calls so it is time-consuming.
2. Shared memory:
   producer-consumer system -> synchronization을 위한 spinlock이 개발되었다.