Kernel Memory Layout on ARM Linux

这是内核自带的文档，讲解ARM芯片的内存是如何布局的！比较简单，对于初学者可以看一下！但要想深入理解Linux内存管理，建议还是找几本好书看看，如深入理解Linux虚拟内存，嵌入系统分析，Linux内核分析及程序设计等； 

Kernel Memory Layout on ARM Linux

Linux在ARM平台上的内存布局

Russell King <rmk@arm.linux.org.uk>
       November 17, 2005 (2.6.15)

This document describes the virtual memory layout which the Linux kernel uses for ARM processors.  It indicates which regions are free for platforms to use, and which are used by generic code.

本文档描述了Linux内核在ARM处理器上的虚拟内存布局。说明了哪些区域是给ARM平台使用的，哪些区域是通用代码使用的。

The ARM CPU is capable of addressing a maximum of 4GB virtual memory space, and this must be shared between user space processes, the kernel, and hardware devices.

ARM系列的CPU最大有4GB的虚拟内存空间寻址能力，但它必须用户空间，内存以及硬件设备共享地址空间。

As the ARM architecture matures, it becomes necessary to reserve certain regions of VM space for use for new facilities; therefore this document may reserve more VM space over time.

在ARM架构成熟的同时，它也开始需要在VM空间保留明确的区域，让用户使用更简单，因此，随着时间的流逝，本文档可能保留了更多的VM空间。

Start（开始地址）  End（结束地址）  Use（使用说明）
--------------------------------------------------------------------------
ffff8000 ffffffff copy_user_page / clear_user_page use. For SA11xx and Xscale, this is used to setup a minicache mapping.

copy_user_page/clear_user_page使用，对于SA11XX和Xscal，它用于设置机器映射。

ffff1000 ffff7fff Reserved.
    Platforms must not use this address range.　保留，ARM平台一定不使用这个区间。

ffff0000 ffff0fff CPU vector page.　CPU向量表
    The CPU vectors are mapped here if the CPU supports vector relocation (control register V bit.)

　　如果CPU支持向量重定向（控制寄存器的V位），则CPU向量被映射到这里。

ffc00000 fffeffff DMA memory mapping region.  Memory returned by the dma_alloc_xxx functions will be dynamically mapped here.

　　DMA内存映射区间。由dma_ammoc_xxx系列函数返回的内存会动态的映射到这里。

ff000000 ffbfffff Reserved for future expansion of DMA　mapping region.

　　保留，用于以后的DMA扩展映射区间。

VMALLOC_END feffffff Free for platform use, recommended.　建议平台保留。
    VMALLOC_END must be aligned to a 2MB boundary.　VMALLOC_END必须在2MB的边界上对齐。　　

VMALLOC_START VMALLOC_END-1 vmalloc() / ioremap() space. vmalloc
    Memory returned by vmalloc/ioremap will be dynamically placed in this region. VMALLOC_START may be based upon the value of the high_memory variable.

　　vmalloc/ioremap 函数返回的内存会被动态的放到该区间。VMALLOC_START 可能的值可能会基于high_memory变量的值而有所不同。

PAGE_OFFSET high_memory-1 Kernel direct-mapped RAM region.　内核直接映射内存区间。
    This maps the platforms RAM, and typically maps all platform RAM in a 1:1 relationship.

　　它映射平台的RAM，通常所有的平台都使用1：1的映射关系。

TASK_SIZE PAGE_OFFSET-1 Kernel module space    Kernel modules inserted via insmod are
    placed here using dynamic mappings.

　　内核模块空间，内核模块通过insmod命令加载，会动态的映射到这里。

00001000 TASK_SIZE-1 User space mappings　用户空间
    Per-thread mappings are placed here via
    the mmap() system call.每个进程通过mmap系统调用的映射放到这里。

00000000 00000fff CPU vector page / null pointer trap CPU向量表，NULL指针陷井
    CPUs which do not support vector remapping place their vector page here.  NULL pointer dereferences by both the kernel and user space are also caught via this mapping.　

　不支持向量重映射的CPU的向量表被映射到这里。用户空间和内核态的NULL指针引通过这个映射可以被捕获。　

Please note that mappings which collide with the above areas may result
in a non-bootable kernel, or may cause the kernel to (eventually) panic
at run time.

请注意：一些与上面冲突的映射会导致内核无法启动，或者可能在运行时会产生（最终）内核panic !

Since future CPUs may impact the kernel mapping layout, user programs
must not access any memory which is not mapped inside their 0x0001000
to TASK_SIZE address range.  If they wish to access these areas, they
must set up their own mappings using open() and mmap().

不管CPU特性会是否与内核的映射布局冲突，用户程序在内部没有映射0x0001000到TASK_SIZE之间的地址空间时，必须不能访问这里面的内存。如果他们想这样做，那么他们必须通过open和mmap来创建自己的映射。