mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Keir Fraser <keir.fraser@eu.citrix.com>
To: "H. Peter Anvin" <hpa@zytor.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>,
	Ian Campbell <Ian.Campbell@eu.citrix.com>,
	Jeremy Fitzhardinge <jeremy@goop.org>,
	the arch/x86 maintainers <x86@kernel.org>,
	Suresh Siddha <suresh.b.siddha@intel.com>,
	"Linux@zytor.com" <Linux@zytor.com>,
	"Pallipadi, Venkatesh" <venkatesh.pallipadi@intel.com>,
	Kernel Mailing List <linux-kernel@vger.kernel.org>,
	Jeremy Fitzhardinge <Jeremy.Fitzhardinge@citrix.com>,
	Xen-devel <xen-devel@lists.xensource.com>,
	Jan Beulich <jbeulich@novell.com>, Ingo Molnar <mingo@elte.hu>,
	Lu <yinghai@kernel.org>, Thomas Gleixner <tglx@linutronix.de>,
	Arjan van de Ven <arjan@infradead.org>
Subject: Re: [Xen-devel] Re: [PATCH 06/16] xen: disable PAT
Date: Sun, 10 May 2009 10:21:48 +0100	[thread overview]
Message-ID: <C62C5B3C.AB1A%keir.fraser@eu.citrix.com> (raw)
In-Reply-To: <4A05CD29.6050505@zytor.com>

On 09/05/2009 19:36, "H. Peter Anvin" <hpa@zytor.com> wrote:

>> My suggestion is that Xen changes its PAT[1] from WT to WC. AFAIK only
>> suse's Linux port is currently using Xen's PAT support, and it does not
>> actually use the WT cache attribute. Then Xen's PAT[0-3] would match what
>> Linux expects, and Linux apparently does not risk using PAT[4-7] because of
>> processor errata so their values are irrelevant. Linux can check that it is
>> running on Xen with suitable PAT setup by reading MSR_IA32_CR_PAT and
>> checking PAT[0-3], and only enable PAT usage in case of a match.
> 
> There exists at least one processor erratum where the CPU will use
> PAT[4-7] when the user requested PAT[0-3].  For those CPUs, it is unsafe
> for *any* OS to have PAT[4-7] != PAT[0-3].

Would it be unreasonable for us to disable PAT on such processors?

An alternative would be to set up PAT according to a new elfnote in the dom0
kernel image, and then only expose the PAT feature flag to domU kernels that
have matching PAT configuration. No elfnote would mean use Xen's existing
PAT setup (or if that's very dangerous then disable PAT altogether, perhaps
dependent on CPU model/stepping?).

 -- Keir



  reply	other threads:[~2009-05-10  9:22 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-05-07 21:10 [GIT PULL] xen: core dom0 support Jeremy Fitzhardinge
2009-05-07 21:10 ` [PATCH 01/16] xen dom0: Make hvc_xen console work for dom0 Jeremy Fitzhardinge
2009-05-07 21:10 ` [PATCH 02/16] xen dom0: Initialize xenbus " Jeremy Fitzhardinge
2009-05-07 21:10 ` [PATCH 03/16] xen dom0: Set up basic IO permissions " Jeremy Fitzhardinge
2009-05-07 21:10 ` [PATCH 04/16] xen dom0: Add support for the platform_ops hypercall Jeremy Fitzhardinge
2009-05-07 21:10 ` [PATCH 05/16] xen mtrr: Add mtrr_ops support for Xen mtrr Jeremy Fitzhardinge
2009-05-08 11:17   ` Ingo Molnar
2009-05-07 21:10 ` [PATCH 06/16] xen: disable PAT Jeremy Fitzhardinge
2009-05-08 11:14   ` Ingo Molnar
2009-05-08 12:38     ` Ian Campbell
2009-05-08 14:50       ` Alan Cox
2009-05-09  8:04         ` [Xen-devel] " Keir Fraser
2009-05-09 18:36           ` H. Peter Anvin
2009-05-10  9:21             ` Keir Fraser [this message]
2009-05-10 13:29               ` Alan Cox
2009-05-10 17:34                 ` Keir Fraser
2009-05-10 17:55                 ` H. Peter Anvin
2009-05-11 11:10                   ` Ingo Molnar
2009-05-07 21:10 ` [PATCH 07/16] xen/dom0: use _PAGE_IOMAP in ioremap to do machine mappings Jeremy Fitzhardinge
2009-05-07 21:10 ` [PATCH 08/16] xen/dom0: Use host E820 map Jeremy Fitzhardinge
2009-05-07 21:10 ` [PATCH 09/16] xen: implement XENMEM_machphys_mapping Jeremy Fitzhardinge
2009-05-07 21:10 ` [PATCH 10/16] xen: clear reserved bits in l3 entries given in the initial pagetables Jeremy Fitzhardinge
2009-05-07 21:10 ` [PATCH 11/16] xen: allow enable use of VGA console on dom0 Jeremy Fitzhardinge
2009-05-07 21:10 ` [PATCH 12/16] xen/dom0: add XEN_DOM0 config option Jeremy Fitzhardinge
2009-05-07 21:10 ` [PATCH 13/16] x86: make /dev/mem mappings _PAGE_IOMAP Jeremy Fitzhardinge
2009-05-07 21:10 ` [PATCH 14/16] paravirtualize IO permission bitmap Jeremy Fitzhardinge
2009-05-07 21:10 ` [PATCH 15/16] x86: don't need "changed" parameter for set_io_bitmap() Jeremy Fitzhardinge
2009-05-07 21:10 ` [PATCH 16/16] xen: checkpatch cleanups Jeremy Fitzhardinge

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=C62C5B3C.AB1A%keir.fraser@eu.citrix.com \
    --to=keir.fraser@eu.citrix.com \
    --cc=Ian.Campbell@eu.citrix.com \
    --cc=Jeremy.Fitzhardinge@citrix.com \
    --cc=Linux@zytor.com \
    --cc=alan@lxorguk.ukuu.org.uk \
    --cc=arjan@infradead.org \
    --cc=hpa@zytor.com \
    --cc=jbeulich@novell.com \
    --cc=jeremy@goop.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=suresh.b.siddha@intel.com \
    --cc=tglx@linutronix.de \
    --cc=venkatesh.pallipadi@intel.com \
    --cc=x86@kernel.org \
    --cc=xen-devel@lists.xensource.com \
    --cc=yinghai@kernel.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

Powered by JetHome