xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: Keir Fraser <keir.xen@gmail.com>
To: Gianni Tedesco <gianni.tedesco@citrix.com>,
	Xen Devel <xen-devel@lists.xensource.com>
Cc: Tim Deegan <Tim.Deegan@eu.citrix.com>, Keir Fraser <keir.fraser@xen.org>
Subject: Re: xc: error: xc_machphys_mfn_list: 83 != 129 when suspending 32GB PV DomU
Date: Fri, 11 Mar 2011 19:21:35 +0000	[thread overview]
Message-ID: <C9A026BF.14A37%keir.xen@gmail.com> (raw)
In-Reply-To: <1299869540.17229.72.camel@qabil.uk.xensource.com>

On 11/03/2011 18:52, "Gianni Tedesco" <gianni.tedesco@citrix.com> wrote:

> Further debugging reveals the variables are set as such:
>  (XEN) compat_machine_to_phys_mapping = 18446606377058041856
>  (XEN) max_page = 67272704
>  (XEN) MACH2PHYS_COMPAT_NR_ENTRIES(current->domain) = 43515904
>  (XEN) RDWR_COMPAT_MPT_VIRT_START = 18446606377058041856
>  (XEN) RDWR_COMPAT_MPT_VIRT_END = 18446606378131783680
>  (XEN) limit = 18446606377232105472, (1 << L2_PAGETABLE_SHIFT) = 2097152
> 
> Could it be that the compat mach-to-phys conversion table size of 1GB is
> too small?

It is insufficient to cover all of the system's memory. The reason for the
limit is that a 1GB M2P table is all that is reasonable to map into a 32-bit
domain's address space while still leaving space for the guest's own
mappings.

We could make the compat m2p bigger solely for mapping by dom0 when doing
save/restore? However, you'd likely still soon hit the mmap limit for the
save/restore process in dom0. It might extend the lifetime of your 32-bit
dom0 for a bit longer however. Long enough for the lightweight HVM container
for PV guests work to get checked in and make our PV 64-bit guest
performance better.

 -- Keir

> Or that there exists some other arbitrary limit on the size
> of domains that can be suspended [when using a 32bit dom0] ?

  reply	other threads:[~2011-03-11 19:21 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-03-11 18:52 xc: error: xc_machphys_mfn_list: 83 != 129 when suspending 32GB PV DomU Gianni Tedesco
2011-03-11 19:21 ` Keir Fraser [this message]
2011-03-14 10:20   ` Ian Campbell
2011-03-14 15:05     ` [PATCH]: Allow tools to map arbitrarily large machphys_mfn_list on 32bit dom0 Gianni Tedesco
2011-03-14 15:08       ` Keir Fraser
2011-03-14 15:11       ` Ian Campbell
2011-03-14 15:19         ` Gianni Tedesco
2011-03-14 15:55           ` Jan Beulich
2011-03-14 16:03             ` Ian Campbell
2011-03-14 16:22               ` Jan Beulich
2011-03-14 16:33                 ` Ian Campbell
2011-03-14 16:54                   ` Keir Fraser
2011-03-14 17:00                     ` Jan Beulich
2011-03-14 17:09                       ` Gianni Tedesco
2011-03-14 16:58                   ` Jan Beulich
2011-03-14 17:11                     ` Gianni Tedesco
2011-03-15 15:00                     ` Ian Campbell
2011-03-15 15:09                       ` Jan Beulich
2011-03-15 15:14                         ` Ian Campbell

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=C9A026BF.14A37%keir.xen@gmail.com \
    --to=keir.xen@gmail.com \
    --cc=Tim.Deegan@eu.citrix.com \
    --cc=gianni.tedesco@citrix.com \
    --cc=keir.fraser@xen.org \
    --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).