From: 박은병 <peb1611@gmail.com>
To: xen-devel@lists.xensource.com, brendan@cs.ubc.ca
Subject: page table pages with checkpoint functionality
Date: Tue, 11 May 2010 21:04:17 +0900 [thread overview]
Message-ID: <AANLkTimy4j3iFXmKfrNIJoZUTrvRZvmA2joUlny4VxLu@mail.gmail.com> (raw)
[-- Attachment #1.1: Type: text/plain, Size: 1262 bytes --]
Hi, all
I am analyzing the xen source about checkpoint functionality
(tools/libxc/xc_domain_save.c, xc_domain_restore.c)
during saving page table pages to disk or another machine( in the case of
live migration), those pages should be canonicalized,
which means that references to actual mfns are replaced with references to
the corresponding pfns
I can understand that, but I could see the another point that I can't
understand
After printing out the page table entries, I found that most of page table
entries has _PAGE_DIRTY, _PAGE_ACCESSED, _PAGE_RW bit.
As soon as guest OS's booting is completed, I checkpointed the guest OS.
(command is xm save [domainU])
L1 pagetable pages:341
L2 pagetable pages:71
L3 pagetable pages:19
L4 pagetable pages:0
(the number of L4 is 0 because of I have used PAE enabled xen)
As futher investigating, L1 page table entries,
page table entries that have _PAGE_DIRTY :131951
page table entries that have _PAGE_ACCESSED :133960
page table entries that have _PAGE_RW :129787
In PAE enabled level 3 paging, each L1 page table has 512 entries, so total
entries 341*512 = 174592
70~80% ptes have those bits
I make sense of many _PAGE_ACCESSED bits.
but, why do those pages have another 2 bits??
Thanks.
--
Eunbyung Park
[-- Attachment #1.2: Type: text/html, Size: 1382 bytes --]
[-- Attachment #2: Type: text/plain, Size: 138 bytes --]
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel
reply other threads:[~2010-05-11 12:04 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=AANLkTimy4j3iFXmKfrNIJoZUTrvRZvmA2joUlny4VxLu@mail.gmail.com \
--to=peb1611@gmail.com \
--cc=brendan@cs.ubc.ca \
--cc=xen-devel@lists.xensource.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).