Pipe

Basic Concepts#

A pipe can use the output of one command as the input of the next command, represented by the symbol | in the command-line environment, which is a form of one-way communication between processes, implemented based on file descriptors.

Working Principle#

A pipe creates a memory buffer, and two processes (commands) communicate through this buffer.

Creating a Pipe#

When creating a pipe using |, the Shell calls pipe() to create the pipe, which includes two file descriptors.

• Write-end file descriptor: A.
• Read-end file descriptor: B.

Command Connection#

• Left command: Redirects the standard output (file descriptor 1) to the write-end A of the pipe.
• Right command: Redirects the standard input (file descriptor 0) to the read-end B of the pipe.

Data Flow#

• Real-time data transmission: The commands on both sides execute in parallel, without waiting for the previous command to finish.
• No temporary files: Data is transmitted in memory.
• Buffer size: Depends on the system.

Buffer Size#

cat /proc/sys/fs/pipe-max-size

1048576

Limitations#

• One-way communication: Two-way communication requires the use of other mechanisms, such as named pipes (FIFO) or sockets.
• Buffer size: If the buffer is full, the upstream command will pause.