1 Kernel Memory Layout on ARM Linux
3 Russell King <rmk@arm.linux.org.uk>
4 November 17, 2005 (2.6.15)
6 This document describes the virtual memory layout which the Linux
7 kernel uses for ARM processors. It indicates which regions are
8 free for platforms to use, and which are used by generic code.
10 The ARM CPU is capable of addressing a maximum of 4GB virtual memory
11 space, and this must be shared between user space processes, the
12 kernel, and hardware devices.
14 As the ARM architecture matures, it becomes necessary to reserve
15 certain regions of VM space for use for new facilities; therefore
16 this document may reserve more VM space over time.
19 --------------------------------------------------------------------------
20 ffff8000 ffffffff copy_user_page / clear_user_page use.
21 For SA11xx and Xscale, this is used to
22 setup a minicache mapping.
24 ffff4000 ffffffff cache aliasing on ARMv6 and later CPUs.
26 ffff1000 ffff7fff Reserved.
27 Platforms must not use this address range.
29 ffff0000 ffff0fff CPU vector page.
30 The CPU vectors are mapped here if the
31 CPU supports vector relocation (control
34 fffe0000 fffeffff XScale cache flush area. This is used
35 in proc-xscale.S to flush the whole data
36 cache. Free for other usage on non-XScale.
38 fff00000 fffdffff Fixmap mapping region. Addresses provided
39 by fix_to_virt() will be located here.
41 ffc00000 ffefffff DMA memory mapping region. Memory returned
42 by the dma_alloc_xxx functions will be
43 dynamically mapped here.
45 ff000000 ffbfffff Reserved for future expansion of DMA
48 VMALLOC_END feffffff Free for platform use, recommended.
49 VMALLOC_END must be aligned to a 2MB
52 VMALLOC_START VMALLOC_END-1 vmalloc() / ioremap() space.
53 Memory returned by vmalloc/ioremap will
54 be dynamically placed in this region.
55 VMALLOC_START may be based upon the value
56 of the high_memory variable.
58 PAGE_OFFSET high_memory-1 Kernel direct-mapped RAM region.
59 This maps the platforms RAM, and typically
60 maps all platform RAM in a 1:1 relationship.
62 TASK_SIZE PAGE_OFFSET-1 Kernel module space
63 Kernel modules inserted via insmod are
64 placed here using dynamic mappings.
66 00001000 TASK_SIZE-1 User space mappings
67 Per-thread mappings are placed here via
68 the mmap() system call.
70 00000000 00000fff CPU vector page / null pointer trap
71 CPUs which do not support vector remapping
72 place their vector page here. NULL pointer
73 dereferences by both the kernel and user
74 space are also caught via this mapping.
76 Please note that mappings which collide with the above areas may result
77 in a non-bootable kernel, or may cause the kernel to (eventually) panic
80 Since future CPUs may impact the kernel mapping layout, user programs
81 must not access any memory which is not mapped inside their 0x0001000
82 to TASK_SIZE address range. If they wish to access these areas, they
83 must set up their own mappings using open() and mmap().