• The memory mapping 在 LinuxVery important in operating systems because it involves efficient cross-process communication & file manipulation
• Today, I’m going to detail a core concept in operating systems: memory mapping

Associate one virtual memory region in the process with one object on disk so that there is a mapping between the two

1. The above mapping procedure = initializes the virtual memory region
2. Once the virtual memory area is initialized, you are swapped around in swap space
3. The mapped objects are called shared objects (normal files/anonymous files)

If the preceding mapping relationship exists, it has the following characteristics

• The virtual memory area of multiple processes has been mapped to a shared object
• If one process writes data to the virtual area
• It is also visible to the process that also maps the shared object to its own virtual memory region

The schematic diagram is as follows

1. Assume that the virtual memory regions of processes 1 and 2 are mapped to the same shared object.
2. When process 1 writes to its virtual memory region, it is also mapped to the virtual memory region in process 2

• The implementation process of memory mapping is mainly throughLinuxSystem call functions under the system:Mmap ()
• This function creates a virtual memory area and maps shared objects
• Its function prototype, specific use & internal process is as follows

Void *mmap(void *start, size_t length, int prot, int flags, int fd, off_t offset); */ mmap(NULL, MAP_SIZE, PROT_READ, */ mmap(NULL, MAP_SIZE, PROT_READ, */ mmap(NULL, MAP_SIZE, PROT_READ, *)) MAP_PRIVATE, fd, 0); /** * Internal principles * Step 1: Create a virtual memory area * Step 2: implement address mapping, that is, the virtual address space of the process ->> Shared object * Note: * a. At this point, the virtual address does not have any data associated with the file, only to establish a mapping relationship * b. When one of the processes writes to virtual memory, the data is truly visible */Copy the code

• Improves data read, write & transfer time performance
  1. The number of data copies is reduced
  2. Efficient interaction between user-space and kernel space (direct interaction through mapped regions)
  3. Memory reads and writes instead of I/O reads and writes
• Improve memory utilization: Through virtual memory & shared objects

In Linux, according to the essential principles & characteristics of memory mapping, its application scenarios are as follows:

1. Implement memory sharing: such as cross-process communication
2. Improve data read/write efficiency: for example, file read/write operations

In the following sections, I’ll detail an example of memory mapping applied to cross-process communication & file operations

• The traditionalLinuxThe process for operating system files is as follows

• A memory-mapped file read/write operation was used

As can be seen from the above: the use of memory mapping file read/write mode is more efficient, the best performance!

• Traditional cross-process communication

• Cross-process communication using memory mapping

As you can see from the above, cross-process communication using memory mapping is the most efficient and performs best!

• This article provides a comprehensive overview of memory mapping in Linux

• I will continue to cover the basics of programming development. If you are interested, please follow the development notes for Carson_Ho