Lecture 5 - Inter-Process Communication

Message Passing

• high-overhead: one syscall per communication operation

Why syscall? Because we need to switch from user mode to kernel mode to access the kernel's data structures.

We have to implement two calls

• send(Q, message): send a message to process Q
• recv(Q, message): recv a message from process Q

Direct Communication

Processes must name each other explicitly:

• send (P, message) – send a message to process P
• receive(Q, message) – receive a message from process Q
• Between each pair there exists exactly one link

Drawbacks:

• Need so many links

Indirect Communication

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

• 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

Synchronization

Buffering

Queue of messages attached to the link.

Implemented in one of three ways

1. Zero capacity – no messages are queued on a link. Sender must wait for receiver
2. Bounded capacity – finite length of n messages. Sender must wait if link full
3. Unbounded capacity – infinite length. Sender never waits

Shared Memory

• One process creates a shared memory segment
• Processes can then “attach” it to their address spaces

Note that this is really contrary to the memory protection idea central to multi-programming!

• Processes communicate by reading/writing to the shared memory region

1. They are responsible for not stepping on each other’s toes
2. The OS is not involved at all

shmget(), shmat(), shmdt(), shmctl()

• Disadvantages 1. Processes must coordinate access to shared memory 2. No protection from other processes: Once another process has the id of the shared memory segment, it can attach to it and read/write to it

Signals

In Lecture 4, we discussed how a process can be terminated by another process. Signals are a way to do this.

Pipes

Ordinary pipes

Cannot be accessed from outside the process that created it.

Typically, a parent process creates a pipe and uses it to communicate with a child process that it created.

Named pipes

• Named Pipes are more powerful than ordinary pipes
• Communication is bidirectional
• No parent-child relationship is necessary between the communicating processes
• Several processes can use the named pipe for communication
• Provided on both UNIX and Windows systems

ls | grep foo

What will shell do ?

1. fork() - Shell 首先创建一个子进程来执行 ls 命令。

2. 创建管道（pipe） - 在 fork() 之后，shell 创建一个管道，管道有两个文件描述符：

   • 写端（供 ls 使用）

   • 读端（供 grep 使用）

3. 子进程1（ls） - 在子进程中：

   • 关闭标准输出（stdout），将其重定向到管道的写端。

   • 关闭管道的读端（子进程不需要读）。

   • 使用 exec() 替换进程镜像，执行 ls 命令。

4. fork() - Shell 创建第二个子进程来运行 grep foo。

5. 子进程2（grep） - 在第二个子进程中：

   • 关闭标准输入（stdin），将其重定向到管道的读端。

   • 关闭管道的写端（子进程不需要写）。

   • 使用 exec() 替换进程镜像，执行 grep foo 命令。 6. 父进程（Shell）：

   • 关闭管道的写端和读端（父进程不直接参与管道的读写）。

   • 等待两个子进程结束。

6. ls 写入管道 - ls 命令的输出通过管道写入。

7. grep 从管道读取 - grep foo 从管道的读端接收数据，进行匹配。
8. ls 完成 - ls 命令完成执行并退出。
9. grep 完成 - grep foo 完成过滤操作，输出匹配结果后退出。

Client-Server Communication

Sockets

Sockets are the most widely used communication mechanism for client-server communication.

• A socket is an endpoint for communication
• A socket is identified by an IP address concatenated with a port number

RPC, RMI......

---

最后更新: 2025年1月6日 21:18:56
创建日期: 2024年12月27日 21:05:43