To elaborate further, PTE is 64-bit (for 32-bit or 64-bit OS) only under 3 level pagetable implementation: Look at arch/x86/include/asm/pgtable-3level_types.h: typedef u64 pteval_t; typedef union { struct { unsigned long pte_low, pte_high; }; pteval_t pte; } pte_t; On Mon, Oct 3, 2011 at 12:05 PM, Peter Teoh <htmldeveloper@gmail.com> wrote:
Basically the concept of 'dirty' is because the memory is file-backed, ie, everytime you changed the memory, it has to be flushed to the file. In Xen the dirty bit is also used to signify the page being modified by the guest, and therefore certain operation like sync-ing between host and guest, or inter-guest sync-ing is possible.
It is just a bit defined in the 64-bit PTE (defined in include/linux/page-flags.h), and to query it for each page in the system is via /proc/kpageflags, internally how that is displayed is implemented in fs/proc/page.c (kernel source).
There is also a tool to query kpageflags:
http://lwn.net/Articles/332300/
On Fri, Sep 30, 2011 at 12:16 AM, Prateek Sharma <prateek3.14@gmail.com> wrote:
Hi all, What would be a good way to get a list of all pages(lets say pfns) dirtied in a particular time interval ? I am reading about MMU_notifiers but i think a simpler, light-weight solution might work.. I came across the page-dirtied bits...are those saved somewhere by the kernel on context-switches? Thanks!
_______________________________________________ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
-- Regards, Peter Teoh
-- Regards, Peter Teoh