All of lore.kernel.org
 help / color / mirror / Atom feed
From: Julien Grall <julien.grall@arm.com>
To: Feng Kan <fkan@apm.com>
Cc: xen-devel@lists.xenproject.org, sstabellini@kernel.org
Subject: Re: [XEN-devel] arm64: fix incorrect pa_range_info table to support 42 bit PA systems.
Date: Mon, 15 May 2017 18:14:04 +0100	[thread overview]
Message-ID: <2dfccbc4-0001-f686-ee03-bd79ae07a975@arm.com> (raw)
In-Reply-To: <CAL85gmAr-UYp3YteoPVV=0Ck+bssdztE8BiiP=xa8nvFta4srg@mail.gmail.com>

Hi,

On 15/05/17 18:11, Feng Kan wrote:
> On Mon, May 15, 2017 at 7:53 AM, Julien Grall <julien.grall@arm.com> wrote:
>> Hello Feng,
>>
>> On 13/05/17 01:26, Feng Kan wrote:
>>>
>>> The pa_range_info table contain incorrect root_order and t0sz which
>>> prevent 42 bit PA systems from booting dom0.
>>
>>
>> As I mentioned in the previous thread [1], this is not a bug. What you
>> configure below is the stage-2 page table and not the hypervisor page-table.
>>
>> It is perfectly fine to expose less IPA (Intermediate Physical Address) bits
>> than the number of PA (Physical Address) bits as long as all the address
>> wired are below 40 bits (assumption made by the patch who added this code).
>> Does your hardware have devices/RAM above 40 bits?
> Yes,  the APM X-Gene series have all been 42 bit PA systems.
> Particularly X-Gene 3, which
> has its PCIe0 all the way up starting at 41 bit.

Thank you for the information.

> If so, then you need to
>> mention in the commit message.
> I will be more clear in the commit message.
>>
>> This bring another question, now you will allocate 8 pages by default for
>> both DOM0 and guests. Exposing 42 bits IPA to a guest does not sound
>> necessary, so we would waste memory here. How are you going to address that?
> To be honest, I hadn't thought of that. I had assume system such as
> these would have
> plenty of memory. I will take a look regarding this. If you have any
> suggestions that would
> be greatly appreciated

I am not totally against using 8 pages. Although a TODO will be useful 
in the code and the commit message.

Cheers,

-- 
Julien Grall

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

      reply	other threads:[~2017-05-15 17:14 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-05-13  0:26 [XEN-devel] arm64: fix incorrect pa_range_info table to support 42 bit PA systems Feng Kan
2017-05-15 14:53 ` Julien Grall
2017-05-15 17:11   ` Feng Kan
2017-05-15 17:14     ` Julien Grall [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=2dfccbc4-0001-f686-ee03-bd79ae07a975@arm.com \
    --to=julien.grall@arm.com \
    --cc=fkan@apm.com \
    --cc=sstabellini@kernel.org \
    --cc=xen-devel@lists.xenproject.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.