2 Virtual memory map with 4 level page tables:
4 0000000000000000 - 00007fffffffffff (=47 bits) user space, different per mm
5 hole caused by [47:63] sign extension
6 ffff800000000000 - ffff87ffffffffff (=43 bits) guard hole, reserved for hypervisor
7 ffff880000000000 - ffffc7ffffffffff (=64 TB) direct mapping of all phys. memory
8 ffffc80000000000 - ffffc8ffffffffff (=40 bits) hole
9 ffffc90000000000 - ffffe8ffffffffff (=45 bits) vmalloc/ioremap space
10 ffffe90000000000 - ffffe9ffffffffff (=40 bits) hole
11 ffffea0000000000 - ffffeaffffffffff (=40 bits) virtual memory map (1TB)
13 ffffec0000000000 - fffffbffffffffff (=44 bits) kasan shadow memory (16TB)
15 fffffe0000000000 - fffffe7fffffffff (=39 bits) LDT remap for PTI
16 fffffe8000000000 - fffffeffffffffff (=39 bits) cpu_entry_area mapping
17 ffffff0000000000 - ffffff7fffffffff (=39 bits) %esp fixup stacks
19 ffffffef00000000 - fffffffeffffffff (=64 GB) EFI region mapping space
21 ffffffff80000000 - ffffffff9fffffff (=512 MB) kernel text mapping, from phys 0
22 ffffffffa0000000 - [fixmap start] (~1526 MB) module mapping space (variable)
23 [fixmap start] - ffffffffff5fffff kernel-internal fixmap range
24 ffffffffff600000 - ffffffffff600fff (=4 kB) legacy vsyscall ABI
25 ffffffffffe00000 - ffffffffffffffff (=2 MB) unused hole
27 Virtual memory map with 5 level page tables:
29 0000000000000000 - 00ffffffffffffff (=56 bits) user space, different per mm
30 hole caused by [56:63] sign extension
31 ff00000000000000 - ff0fffffffffffff (=52 bits) guard hole, reserved for hypervisor
32 ff10000000000000 - ff8fffffffffffff (=55 bits) direct mapping of all phys. memory
33 ff90000000000000 - ff9fffffffffffff (=52 bits) LDT remap for PTI
34 ffa0000000000000 - ffd1ffffffffffff (=54 bits) vmalloc/ioremap space (12800 TB)
35 ffd2000000000000 - ffd3ffffffffffff (=49 bits) hole
36 ffd4000000000000 - ffd5ffffffffffff (=49 bits) virtual memory map (512TB)
38 ffdf000000000000 - fffffc0000000000 (=53 bits) kasan shadow memory (8PB)
40 fffffe8000000000 - fffffeffffffffff (=39 bits) cpu_entry_area mapping
41 ffffff0000000000 - ffffff7fffffffff (=39 bits) %esp fixup stacks
43 ffffffef00000000 - fffffffeffffffff (=64 GB) EFI region mapping space
45 ffffffff80000000 - ffffffff9fffffff (=512 MB) kernel text mapping, from phys 0
46 ffffffffa0000000 - fffffffffeffffff (1520 MB) module mapping space
47 [fixmap start] - ffffffffff5fffff kernel-internal fixmap range
48 ffffffffff600000 - ffffffffff600fff (=4 kB) legacy vsyscall ABI
49 ffffffffffe00000 - ffffffffffffffff (=2 MB) unused hole
51 Architecture defines a 64-bit virtual address. Implementations can support
52 less. Currently supported are 48- and 57-bit virtual addresses. Bits 63
53 through to the most-significant implemented bit are sign extended.
54 This causes hole between user space and kernel addresses if you interpret them
57 The direct mapping covers all memory in the system up to the highest
58 memory address (this means in some cases it can also include PCI memory
61 vmalloc space is lazily synchronized into the different PML4/PML5 pages of
62 the processes using the page fault handler, with init_top_pgt as
65 We map EFI runtime services in the 'efi_pgd' PGD in a 64Gb large virtual
66 memory window (this size is arbitrary, it can be raised later if needed).
67 The mappings are not part of any other kernel PGD and are only available
68 during EFI runtime calls.
70 Note that if CONFIG_RANDOMIZE_MEMORY is enabled, the direct mapping of all
71 physical memory, vmalloc/ioremap space and virtual memory map are randomized.
72 Their order is preserved but their base will be offset early at boot time.