xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: Jaeyong Yoo <jaeyong.yoo@samsung.com>
To: 'Ian Campbell' <Ian.Campbell@citrix.com>
Cc: 'Elena Pyatunina' <e.pyatunina@samsung.com>, xen-devel@lists.xen.org
Subject: Re: [PATCH RFC v2 6/9] xen/arm: Add handling write fault for dirty-page tracing
Date: Thu, 04 Jul 2013 20:47:20 +0900	[thread overview]
Message-ID: <003301ce78ac$3d80c3c0$b8824b40$%yoo@samsung.com> (raw)
In-Reply-To: <1372927595.7184.10.camel@kazak.uk.xensource.com>

> -----Original Message-----
> From: Ian Campbell [mailto:Ian.Campbell@citrix.com]
> Sent: Thursday, July 04, 2013 5:47 PM
> To: Jaeyong Yoo
> Cc: xen-devel@lists.xen.org; 'Elena Pyatunina'
> Subject: Re: [Xen-devel] [PATCH RFC v2 6/9] xen/arm: Add handling write
> fault for dirty-page tracing
> 
> On Thu, 2013-07-04 at 10:02 +0900, Jaeyong Yoo wrote:
> 
> > > I had been considering the option of reusing one of the bits
> > > available for software use in the p2m entries to track the dirty
> > > status and walking the p2m when the toolstack asked for the current
> bitmap instead.
> > >
> > > I think this would have less memory overhead and also would allow us
> > > to be lock free on the fault handling path (by using suitable atomic
> > > exchanges on the p2m entries), as well as avoiding memory allocations
> on that path.
> > >
> > > I think we'd want to establish a linear map of the current guest p2m
> > > for these purposes so we could also avoid all the map_domain_page
> > > stuff in the fault path. I think there is enough virtual address
> > > space left for that even on 32-bit, it's less critical on 64-bit
> > > anyway since we are about to have a direct map of RAM available to us.
> >
> > I think it is a good idea to have an ever-mapping linear dirty-bit map
> > for guest p2m.
> 
> By Linear map I meant slotting the p2m into the hypervisor's own page
> tables (I believe the relevant p2m and pte bits in the entries do not
> overlap, so this is possible), such that the p2m table entries are
> available at known virtual addresses and you can locate the L1, L2 and
> L3 entries corresponding to a given guest physical address with just
> arithmetic, i.e. something like:
> http://www.technovelty.org/linux/virtual-linear-page-table.html
> 

Oh it looks interesting. And we can try this. Thanks for the good tip :)

> As opposed to an always mapped version of the dirty bitmap, which I think
> we can sync only when the tools ask.

Got it.

Jaeyong

  reply	other threads:[~2013-07-04 11:47 UTC|newest]

Thread overview: 44+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-07-03  9:15 [PATCH RFC v2 0/9] Preliminary working version of live migration Jaeyong Yoo
2013-07-03  9:15 ` [PATCH RFC v2 1/9] xen-arm: Implement basic save/load for hvm context Jaeyong Yoo
2013-07-03 11:29   ` Stefano Stabellini
2013-07-03 23:51     ` Jaeyong Yoo
2013-07-03  9:15 ` [PATCH RFC v2 2/9] xen/arm: Implement save and restore for gic, vtimer, and ptimer Jaeyong Yoo
2013-07-03 11:29   ` Stefano Stabellini
2013-07-03 23:53     ` Jaeyong Yoo
2013-07-03  9:15 ` [PATCH RFC v2 3/9] xen/arm: Add more registers for saving and restoring vcpu registers Jaeyong Yoo
2013-07-03 11:29   ` Stefano Stabellini
2013-07-03  9:15 ` [PATCH RFC v2 4/9] xen/arm: Implement get_maximum_gpfn hypercall for arm Jaeyong Yoo
2013-07-03 11:29   ` Stefano Stabellini
2013-07-03 11:35     ` Ian Campbell
2013-07-04  0:09       ` Jaeyong Yoo
2013-07-04  8:47         ` Ian Campbell
2013-07-03  9:15 ` [PATCH RFC v2 5/9] xen/arm: Implement modify_returncode Jaeyong Yoo
2013-07-03  9:15 ` [PATCH RFC v2 6/9] xen/arm: Add handling write fault for dirty-page tracing Jaeyong Yoo
2013-07-03 12:10   ` Stefano Stabellini
2013-07-04  0:44     ` Jaeyong Yoo
2013-07-03 12:26   ` Ian Campbell
2013-07-04  1:02     ` Jaeyong Yoo
2013-07-04  7:21       ` Jaeyong Yoo
2013-07-04  8:46       ` Ian Campbell
2013-07-04 11:47         ` Jaeyong Yoo [this message]
2013-07-05  4:42         ` Jaeyong Yoo
2013-07-07 16:53           ` Ian Campbell
2013-07-12  0:54             ` Jaeyong Yoo
2013-07-03  9:15 ` [PATCH RFC v2 7/9] xen/arm: Missing impl of clear_guest_offset macro Jaeyong Yoo
2013-07-03 11:37   ` Stefano Stabellini
2013-07-03 11:57     ` Ian Campbell
2013-07-04  0:09     ` Jaeyong Yoo
2013-07-03  9:15 ` [PATCH RFC v2 8/9] xen/arm: Implement hypercall for dirty page tracing (shadow op) Jaeyong Yoo
2013-07-03 12:38   ` Stefano Stabellini
2013-07-04  1:25     ` Jaeyong Yoo
2013-07-04  8:42       ` Ian Campbell
2013-07-04 10:18         ` Jaeyong Yoo
2013-07-04 10:29           ` Stefano Stabellini
2013-07-04 10:36             ` Ian Campbell
2013-07-10 15:34               ` Eugene Fedotov
2013-07-10 15:39                 ` Ian Campbell
2013-07-11 10:18                   ` Eugene Fedotov
2013-07-11 11:24                     ` Ian Campbell
2013-07-03  9:15 ` [PATCH RFC v2 9/9] xen/arm: Implement toolstack for xl restore/save and migrate Jaeyong Yoo
2013-07-03  9:21 ` [PATCH RFC v2 0/9] Preliminary working version of live migration Ian Campbell
2013-07-03 12:54 ` Stefano Stabellini

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='003301ce78ac$3d80c3c0$b8824b40$%yoo@samsung.com' \
    --to=jaeyong.yoo@samsung.com \
    --cc=Ian.Campbell@citrix.com \
    --cc=e.pyatunina@samsung.com \
    --cc=xen-devel@lists.xen.org \
    /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).