Linux 内存页设计说明

内存页

Linux 内存模型关于页的设计主要有以下几点：

页面大小：Linux 的页面大小通常为 4KB，表示内存分页的最小单元。页面大小对于系统的性能和内存管理有着重要的影响，因为它影响系统中页表项数量的大小以及内存映射的效率。

页表：Linux 内存管理使用了虚拟地址到物理地址的转换机制，涉及到多级页表的设计，以使操作系统和程序能够使用虚拟地址，而不必知道实际的物理内存地址。每个进程都有一个页表，用于保存虚拟地址和物理地址的映射信息。

物理内存管理：Linux 内存管理使用页框来管理物理内存，页框是物理内存的最小单元，大小通常为 4KB。当内存不足时，Linux 内存管理可以使用交换空间（swap space）将不常用的页面暂时存储到硬盘上，释放物理内存空间。

内存分配：Linux 内存管理使用伙伴算法（buddy algorithm）来分配物理内存页框。这个算法将可用内存空间划分成一组大小相等的块，每个块的大小是 2 的幂次方。当需要分配一个块时，算法会找到一个与所需大小最接近的块，然后将其加入到空闲块列表中。分配物理内存的操作需要在操作系统内核模式下进行，因此只有系统内核代码才能分配物理内存。

Page unit

In Linux, the memory management system is based on a paging model, where memory is divided into fixed-size blocks called pages. The page size is typically 4KB or 8KB, depending on the architecture and configuration of the system.
The design of pages in Linux allows for efficient memory allocation and deallocation, as well as memory protection and sharing between processes. When a process requests memory, the kernel allocates one or more pages of memory and maps them to the process’s virtual address space. The process can then access the memory through its virtual address, and the kernel ensures that the virtual memory is mapped to the correct physical memory page.
Pages in Linux are managed using a page table, which is a data structure that keeps track of the mapping between virtual and physical memory pages. The page table is maintained by the kernel and is used to translate virtual memory addresses to physical memory addresses. The page table also includes information about the status of each page, such as whether it is currently in use, free, or swapped out to disk.
In addition to efficient memory allocation and deallocation, the use of pages in Linux also allows for memory protection and sharing between processes. Pages can be marked as read-only or no-execute to prevent unauthorized access or execution. Pages can also be shared between processes, allowing multiple processes to access the same physical memory page.
Overall, the design of pages in Linux is a key component of the memory management system, providing efficient memory allocation and deallocation, memory protection, and sharing between processes.