All of lore.kernel.org
 help / color / mirror / Atom feed
From: Andrew Cooper <andrew.cooper3@citrix.com>
To: Yunbin Wang <wangyunbin1989@gmail.com>
Cc: Xen-devel List <xen-devel@lists.xen.org>
Subject: Re: Memory strategy in xen
Date: Tue, 3 Dec 2013 14:18:51 +0000	[thread overview]
Message-ID: <529DE84B.20407@citrix.com> (raw)
In-Reply-To: <CAEPeN01y7cD9kdFzBcmb+J7FJGL124x=xKNT0HAmimDFMkatSA@mail.gmail.com>


[-- Attachment #1.1: Type: text/plain, Size: 2629 bytes --]

On 03/12/13 13:48, Yunbin Wang wrote:
> Hi Andrew,
> Thanks for your reply.
>
> How does Xen map guest virtual address to machine physical address in
> PV mode?
>
> In PV mode, guests are aware of VMM, and know the actual physical
> memory address assigned to them, so do they map guest virtual address
> to guest physical address first, and then map physical address to
> machine physical, OR they map virtual address to machine physical
> directly?
>
> Thanks,
> Yunbin Wang

Please do not drop xen-devel from CC, and please do not top-post.

Xen does not keep track of guest virtual address - that is the guests
job to do.  Xen maintains the pagetables, but the PV guest has to ask
Xen to update a PTE, and what value to update it to.

PV guests are aware that their physical address space is not contiguous,
and have to take this into account for memory management purposes.

~Andrew

>
>
> 2013/11/18 Andrew Cooper <andrew.cooper3@citrix.com
> <mailto:andrew.cooper3@citrix.com>>
>
>     On 17/11/13 00:26, Yunbin Wang wrote:
>>     Hi all,
>>     I'm a newer to xen. But now I need to do research about live
>>     migration in xen. And now I'm confused with some questions:
>>     1. Is shadowing page used in both paravirtualization and hvm
>>     mode? Since from source code xc_domain_save.c, I see the
>>     following code to get dirty memory bitmap,
>>
>>     frc = xc_shadow_control(
>>                         xch, dom, XEN_DOMCTL_SHADOW_OP_PEEK,
>>     HYPERCALL_BUFFER(to_skip),
>>                         dinfo->p2m_size, NULL, 0, NULL);
>
>     From domctl.h
>
>     /* Log-dirty bitmap operations. */
>      /* Return the bitmap and clean internal copy for next round. */
>     #define XEN_DOMCTL_SHADOW_OP_CLEAN       11
>      /* Return the bitmap but do not modify internal copy. */
>     #define XEN_DOMCTL_SHADOW_OP_PEEK        12
>
>     Shadow and logdirty are related, but not the same.  Shadow is
>     certainly not expected to be used with PV guests (although a PV
>     guest specifying autotranslate will be run with shadowing enabled).
>
>     ~Andrew
>
>>
>>
>>     2. The strategy about memory allocation in xen, I know that dom0
>>     will get all available memory at first, and xen will allocate
>>     memory via balloon strategy to domU.
>>
>>     And now I need to know how balloon works?
>>
>>     Can anyone give me some ideas about questions above.
>>
>>     Thanks,
>>     Yunbin Wang
>>
>>
>>     _______________________________________________
>>     Xen-devel mailing list
>>     Xen-devel@lists.xen.org <mailto:Xen-devel@lists.xen.org>
>>     http://lists.xen.org/xen-devel
>
>


[-- Attachment #1.2: Type: text/html, Size: 6420 bytes --]

[-- Attachment #2: Type: text/plain, Size: 126 bytes --]

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel

      parent reply	other threads:[~2013-12-03 14:18 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-11-17  0:26 Memory strategy in xen Yunbin Wang
2013-11-18 16:39 ` George Dunlap
     [not found] ` <5289F3A1.10000@citrix.com>
     [not found]   ` <CAEPeN01y7cD9kdFzBcmb+J7FJGL124x=xKNT0HAmimDFMkatSA@mail.gmail.com>
2013-12-03 14:18     ` Andrew Cooper [this message]

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=529DE84B.20407@citrix.com \
    --to=andrew.cooper3@citrix.com \
    --cc=wangyunbin1989@gmail.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.