1 =============== 1 ===============
2 Soft-Dirty PTEs 2 Soft-Dirty PTEs
3 =============== 3 ===============
4 4
5 The soft-dirty is a bit on a PTE which helps t 5 The soft-dirty is a bit on a PTE which helps to track which pages a task
6 writes to. In order to do this tracking one sh 6 writes to. In order to do this tracking one should
7 7
8 1. Clear soft-dirty bits from the task's PTE 8 1. Clear soft-dirty bits from the task's PTEs.
9 9
10 This is done by writing "4" into the ``/p 10 This is done by writing "4" into the ``/proc/PID/clear_refs`` file of the
11 task in question. 11 task in question.
12 12
13 2. Wait some time. 13 2. Wait some time.
14 14
15 3. Read soft-dirty bits from the PTEs. 15 3. Read soft-dirty bits from the PTEs.
16 16
17 This is done by reading from the ``/proc/ 17 This is done by reading from the ``/proc/PID/pagemap``. The bit 55 of the
18 64-bit qword is the soft-dirty one. If se 18 64-bit qword is the soft-dirty one. If set, the respective PTE was
19 written to since step 1. 19 written to since step 1.
20 20
21 21
22 Internally, to do this tracking, the writable 22 Internally, to do this tracking, the writable bit is cleared from PTEs
23 when the soft-dirty bit is cleared. So, after 23 when the soft-dirty bit is cleared. So, after this, when the task tries to
24 modify a page at some virtual address the #PF 24 modify a page at some virtual address the #PF occurs and the kernel sets
25 the soft-dirty bit on the respective PTE. 25 the soft-dirty bit on the respective PTE.
26 26
27 Note, that although all the task's address spa 27 Note, that although all the task's address space is marked as r/o after the
28 soft-dirty bits clear, the #PF-s that occur af 28 soft-dirty bits clear, the #PF-s that occur after that are processed fast.
29 This is so, since the pages are still mapped t 29 This is so, since the pages are still mapped to physical memory, and thus all
30 the kernel does is finds this fact out and put 30 the kernel does is finds this fact out and puts both writable and soft-dirty
31 bits on the PTE. 31 bits on the PTE.
32 32
33 While in most cases tracking memory changes by 33 While in most cases tracking memory changes by #PF-s is more than enough
34 there is still a scenario when we can lose sof 34 there is still a scenario when we can lose soft dirty bits -- a task
35 unmaps a previously mapped memory region and t 35 unmaps a previously mapped memory region and then maps a new one at exactly
36 the same place. When unmap is called, the kern 36 the same place. When unmap is called, the kernel internally clears PTE values
37 including soft dirty bits. To notify user spac 37 including soft dirty bits. To notify user space application about such
38 memory region renewal the kernel always marks 38 memory region renewal the kernel always marks new memory regions (and
39 expanded regions) as soft dirty. 39 expanded regions) as soft dirty.
40 40
41 This feature is actively used by the checkpoin 41 This feature is actively used by the checkpoint-restore project. You
42 can find more details about it on http://criu. 42 can find more details about it on http://criu.org
43 43
44 44
45 -- Pavel Emelyanov, Apr 9, 2013 45 -- Pavel Emelyanov, Apr 9, 2013
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.