Based on kernel version 4.16.1. Page generated on 2018-04-09 11:53 EST.
1 2 Virtual memory map with 4 level page tables: 3 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) 12 ... unused hole ... 13 ffffec0000000000 - fffffbffffffffff (=44 bits) kasan shadow memory (16TB) 14 ... unused hole ... 15 vaddr_end for KASLR 16 fffffe0000000000 - fffffe7fffffffff (=39 bits) cpu_entry_area mapping 17 fffffe8000000000 - fffffeffffffffff (=39 bits) LDT remap for PTI 18 ffffff0000000000 - ffffff7fffffffff (=39 bits) %esp fixup stacks 19 ... unused hole ... 20 ffffffef00000000 - fffffffeffffffff (=64 GB) EFI region mapping space 21 ... unused hole ... 22 ffffffff80000000 - ffffffff9fffffff (=512 MB) kernel text mapping, from phys 0 23 ffffffffa0000000 - [fixmap start] (~1526 MB) module mapping space (variable) 24 [fixmap start] - ffffffffff5fffff kernel-internal fixmap range 25 ffffffffff600000 - ffffffffff600fff (=4 kB) legacy vsyscall ABI 26 ffffffffffe00000 - ffffffffffffffff (=2 MB) unused hole 27 28 Virtual memory map with 5 level page tables: 29 30 0000000000000000 - 00ffffffffffffff (=56 bits) user space, different per mm 31 hole caused by [56:63] sign extension 32 ff00000000000000 - ff0fffffffffffff (=52 bits) guard hole, reserved for hypervisor 33 ff10000000000000 - ff8fffffffffffff (=55 bits) direct mapping of all phys. memory 34 ff90000000000000 - ff9fffffffffffff (=52 bits) LDT remap for PTI 35 ffa0000000000000 - ffd1ffffffffffff (=54 bits) vmalloc/ioremap space (12800 TB) 36 ffd2000000000000 - ffd3ffffffffffff (=49 bits) hole 37 ffd4000000000000 - ffd5ffffffffffff (=49 bits) virtual memory map (512TB) 38 ... unused hole ... 39 ffdf000000000000 - fffffc0000000000 (=53 bits) kasan shadow memory (8PB) 40 ... unused hole ... 41 vaddr_end for KASLR 42 fffffe0000000000 - fffffe7fffffffff (=39 bits) cpu_entry_area mapping 43 ... unused hole ... 44 ffffff0000000000 - ffffff7fffffffff (=39 bits) %esp fixup stacks 45 ... unused hole ... 46 ffffffef00000000 - fffffffeffffffff (=64 GB) EFI region mapping space 47 ... unused hole ... 48 ffffffff80000000 - ffffffff9fffffff (=512 MB) kernel text mapping, from phys 0 49 ffffffffa0000000 - fffffffffeffffff (1520 MB) module mapping space 50 [fixmap start] - ffffffffff5fffff kernel-internal fixmap range 51 ffffffffff600000 - ffffffffff600fff (=4 kB) legacy vsyscall ABI 52 ffffffffffe00000 - ffffffffffffffff (=2 MB) unused hole 53 54 Architecture defines a 64-bit virtual address. Implementations can support 55 less. Currently supported are 48- and 57-bit virtual addresses. Bits 63 56 through to the most-significant implemented bit are sign extended. 57 This causes hole between user space and kernel addresses if you interpret them 58 as unsigned. 59 60 The direct mapping covers all memory in the system up to the highest 61 memory address (this means in some cases it can also include PCI memory 62 holes). 63 64 vmalloc space is lazily synchronized into the different PML4/PML5 pages of 65 the processes using the page fault handler, with init_top_pgt as 66 reference. 67 68 We map EFI runtime services in the 'efi_pgd' PGD in a 64Gb large virtual 69 memory window (this size is arbitrary, it can be raised later if needed). 70 The mappings are not part of any other kernel PGD and are only available 71 during EFI runtime calls. 72 73 Note that if CONFIG_RANDOMIZE_MEMORY is enabled, the direct mapping of all 74 physical memory, vmalloc/ioremap space and virtual memory map are randomized. 75 Their order is preserved but their base will be offset early at boot time. 76 77 Be very careful vs. KASLR when changing anything here. The KASLR address 78 range must not overlap with anything except the KASAN shadow area, which is 79 correct as KASAN disables KASLR.