46 lines · plain
1===============2Soft-Dirty PTEs3===============4 5The soft-dirty is a bit on a PTE which helps to track which pages a task6writes to. In order to do this tracking one should7 8 1. Clear soft-dirty bits from the task's PTEs.9 10 This is done by writing "4" into the ``/proc/PID/clear_refs`` file of the11 task in question.12 13 2. Wait some time.14 15 3. Read soft-dirty bits from the PTEs.16 17 This is done by reading from the ``/proc/PID/pagemap``. The bit 55 of the18 64-bit qword is the soft-dirty one. If set, the respective PTE was19 written to since step 1.20 21 22Internally, to do this tracking, the writable bit is cleared from PTEs23when the soft-dirty bit is cleared. So, after this, when the task tries to24modify a page at some virtual address the #PF occurs and the kernel sets25the soft-dirty bit on the respective PTE.26 27Note, that although all the task's address space is marked as r/o after the28soft-dirty bits clear, the #PF-s that occur after that are processed fast.29This is so, since the pages are still mapped to physical memory, and thus all30the kernel does is finds this fact out and puts both writable and soft-dirty31bits on the PTE.32 33While in most cases tracking memory changes by #PF-s is more than enough34there is still a scenario when we can lose soft dirty bits -- a task35unmaps a previously mapped memory region and then maps a new one at exactly36the same place. When unmap is called, the kernel internally clears PTE values37including soft dirty bits. To notify user space application about such38memory region renewal the kernel always marks new memory regions (and39expanded regions) as soft dirty.40 41This feature is actively used by the checkpoint-restore project. You42can find more details about it on http://criu.org43 44 45-- Pavel Emelyanov, Apr 9, 201346