From mboxrd@z Thu Jan  1 00:00:00 1970
From: Andrew Cooper <andrew.cooper3@citrix.com>
Subject: Re: [PATCH] document IOMMU related command line options
Date: Fri, 4 Jul 2014 15:53:23 +0100
Message-ID: <53B6BFE3.8040603@citrix.com>
References: <53B6D9B50200007800020CB5@mail.emea.novell.com>
Mime-Version: 1.0
Content-Type: multipart/mixed; boundary="===============0118481568339499055=="
Return-path: <xen-devel-bounces@lists.xen.org>
Received: from mail6.bemta14.messagelabs.com ([193.109.254.103])
	by lists.xen.org with esmtp (Exim 4.72)
	(envelope-from <Andrew.Cooper3@citrix.com>) id 1X34rh-0001kf-Fo
	for xen-devel@lists.xenproject.org; Fri, 04 Jul 2014 14:53:29 +0000
In-Reply-To: <53B6D9B50200007800020CB5@mail.emea.novell.com>
List-Unsubscribe: <http://lists.xen.org/cgi-bin/mailman/options/xen-devel>,
	<mailto:xen-devel-request@lists.xen.org?subject=unsubscribe>
List-Post: <mailto:xen-devel@lists.xen.org>
List-Help: <mailto:xen-devel-request@lists.xen.org?subject=help>
List-Subscribe: <http://lists.xen.org/cgi-bin/mailman/listinfo/xen-devel>,
	<mailto:xen-devel-request@lists.xen.org?subject=subscribe>
Sender: xen-devel-bounces@lists.xen.org
Errors-To: xen-devel-bounces@lists.xen.org
To: Jan Beulich <JBeulich@suse.com>, xen-devel <xen-devel@lists.xenproject.org>
List-Id: xen-devel@lists.xenproject.org

--===============0118481568339499055==
Content-Type: multipart/alternative;
	boundary="------------000708000409090103090702"

--------------000708000409090103090702
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 7bit

On 04/07/14 15:43, Jan Beulich wrote:
> Signed-off-by: Jan Beulich <jbeulich@suse.com>

Awesome.  This has been high on my docs todo list for a long time.

>
> --- a/docs/misc/xen-command-line.markdown
> +++ b/docs/misc/xen-command-line.markdown
> @@ -667,9 +667,101 @@ debug hypervisor only).
>  
>  ### ioapic\_ack
>  ### iommu
> -### iommu\_inclusive\_mapping
> +> `= List of [ <boolean> | force | required | intremap | qinval | snoop | sharept | dom0-passthrough | dom0-strict | amd-iommu-perdev-intremap | workaround_bios_bug | verbose | debug ]`
> +
> +> Sub-options:
> +
> +> `<boolean>`
> +
> +> Default: `on`
> +
> +>> Control the use of IOMMU(s) in the system.
> +
> +> All other sub-options are of boolean kind and can be prefixed with `no-` to
> +> effect the inverse meaning.
> +
> +> `force` or `required`
> +
> +> Default: `false`
> +
> +>> Don't continue booting unless IOMMU support is found and can be initialized
> +>> successfully.
> +
> +> `intremap`
> +
> +> Default: `true`
> +
> +>> Control the use of interrupt remapping (DMA remapping will always be enabled
> +>> if IOMMU functionality is enabled).
> +
> +> `qinval` (VT-d)
> +
> +> Default: `true`
> +
> +>> Control the use of Queued Invalidation.
> +
> +> `snoop` (Intel)

Can we have consistent use of Intel vs VT-d?  Probably VT-d is better.

Other than this,

Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>

> +
> +> Default: `true`
> +
> +>> Control the use of Snoop Control.
> +
> +> `sharept`
> +
> +> Default: `true`
> +
> +>> Control whether CPU and IOMMU page tables should be shared.
> +
> +> `dom0-passthrough`
> +
> +> Default: `false`
> +
> +>> Control whether to disable DMA remapping for Dom0.
> +
> +> `dom0-strict`
> +
> +> Default: `false`
> +
> +>> Control whether to set up DMA remapping only for the memory Dom0 actually
> +>> got assigned. Implies `no-dom0-passthrough`.
> +
> +> `amd-iommu-perdev-intremap`
> +
> +> Default: `true`
> +
> +>> Control whether to set up interrupt remapping data structures per device
> +>> rather that once for the entire system. Turning this off is making PCI
> +>> device pass-through insecure and hence unsupported.
> +
> +> `workaround_bios_bug` (VT-d)
> +
> +> Default: `false`
> +
> +>> Causes DRHD entries without any PCI discoverable devices under them to be
> +>> ignored (normally IOMMU setup fails if any of the devices listed by a DRHD
> +>> entry aren't PCI discoverable).
> +
> +> `verbose`
> +
> +> Default: `false`
> +
> +>> Increase IOMMU code's verbosity.
> +
> +> `debug`
> +
> +> Default: `false`
> +
> +>> Enable IOMMU debugging code (implies `verbose`).
> +
> +### iommu\_inclusive\_mapping (VT-d)
>  > `= <boolean>`
>  
> +> Default: `false`
> +
> +Use this to work around firmware issues providing correct RMRR entries. Rather
> +than only mapping RAM pages for IOMMU accesses for Dom0, with this option all
> +pages not marked as unusable in the E820 table will get a mapping established.
> +
>  ### irq\_ratelimit
>  > `= <integer>`
>  
>
>
>
>
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xen.org
> http://lists.xen.org/xen-devel


--------------000708000409090103090702
Content-Type: text/html; charset="ISO-8859-1"
Content-Transfer-Encoding: 7bit

<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <div class="moz-cite-prefix">On 04/07/14 15:43, Jan Beulich wrote:<br>
    </div>
    <blockquote cite="mid:53B6D9B50200007800020CB5@mail.emea.novell.com"
      type="cite">
      <pre wrap="">Signed-off-by: Jan Beulich <a class="moz-txt-link-rfc2396E" href="mailto:jbeulich@suse.com">&lt;jbeulich@suse.com&gt;</a></pre>
    </blockquote>
    <br>
    Awesome.&nbsp; This has been high on my docs todo list for a long time.<br>
    <br>
    <blockquote cite="mid:53B6D9B50200007800020CB5@mail.emea.novell.com"
      type="cite">
      <pre wrap="">

--- a/docs/misc/xen-command-line.markdown
+++ b/docs/misc/xen-command-line.markdown
@@ -667,9 +667,101 @@ debug hypervisor only).
 
 ### ioapic\_ack
 ### iommu
-### iommu\_inclusive\_mapping
+&gt; `= List of [ &lt;boolean&gt; | force | required | intremap | qinval | snoop | sharept | dom0-passthrough | dom0-strict | amd-iommu-perdev-intremap | workaround_bios_bug | verbose | debug ]`
+
+&gt; Sub-options:
+
+&gt; `&lt;boolean&gt;`
+
+&gt; Default: `on`
+
+&gt;&gt; Control the use of IOMMU(s) in the system.
+
+&gt; All other sub-options are of boolean kind and can be prefixed with `no-` to
+&gt; effect the inverse meaning.
+
+&gt; `force` or `required`
+
+&gt; Default: `false`
+
+&gt;&gt; Don't continue booting unless IOMMU support is found and can be initialized
+&gt;&gt; successfully.
+
+&gt; `intremap`
+
+&gt; Default: `true`
+
+&gt;&gt; Control the use of interrupt remapping (DMA remapping will always be enabled
+&gt;&gt; if IOMMU functionality is enabled).
+
+&gt; `qinval` (VT-d)
+
+&gt; Default: `true`
+
+&gt;&gt; Control the use of Queued Invalidation.
+
+&gt; `snoop` (Intel)</pre>
    </blockquote>
    <br>
    Can we have consistent use of Intel vs VT-d?&nbsp; Probably VT-d is
    better.<br>
    <br>
    Other than this,<br>
    <br>
    Reviewed-by: Andrew Cooper <a class="moz-txt-link-rfc2396E" href="mailto:andrew.cooper3@citrix.com">&lt;andrew.cooper3@citrix.com&gt;</a><br>
    <br>
    <blockquote cite="mid:53B6D9B50200007800020CB5@mail.emea.novell.com"
      type="cite">
      <pre wrap="">
+
+&gt; Default: `true`
+
+&gt;&gt; Control the use of Snoop Control.
+
+&gt; `sharept`
+
+&gt; Default: `true`
+
+&gt;&gt; Control whether CPU and IOMMU page tables should be shared.
+
+&gt; `dom0-passthrough`
+
+&gt; Default: `false`
+
+&gt;&gt; Control whether to disable DMA remapping for Dom0.
+
+&gt; `dom0-strict`
+
+&gt; Default: `false`
+
+&gt;&gt; Control whether to set up DMA remapping only for the memory Dom0 actually
+&gt;&gt; got assigned. Implies `no-dom0-passthrough`.
+
+&gt; `amd-iommu-perdev-intremap`
+
+&gt; Default: `true`
+
+&gt;&gt; Control whether to set up interrupt remapping data structures per device
+&gt;&gt; rather that once for the entire system. Turning this off is making PCI
+&gt;&gt; device pass-through insecure and hence unsupported.
+
+&gt; `workaround_bios_bug` (VT-d)
+
+&gt; Default: `false`
+
+&gt;&gt; Causes DRHD entries without any PCI discoverable devices under them to be
+&gt;&gt; ignored (normally IOMMU setup fails if any of the devices listed by a DRHD
+&gt;&gt; entry aren't PCI discoverable).
+
+&gt; `verbose`
+
+&gt; Default: `false`
+
+&gt;&gt; Increase IOMMU code's verbosity.
+
+&gt; `debug`
+
+&gt; Default: `false`
+
+&gt;&gt; Enable IOMMU debugging code (implies `verbose`).
+
+### iommu\_inclusive\_mapping (VT-d)
 &gt; `= &lt;boolean&gt;`
 
+&gt; Default: `false`
+
+Use this to work around firmware issues providing correct RMRR entries. Rather
+than only mapping RAM pages for IOMMU accesses for Dom0, with this option all
+pages not marked as unusable in the E820 table will get a mapping established.
+
 ### irq\_ratelimit
 &gt; `= &lt;integer&gt;`
 



</pre>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <pre wrap="">_______________________________________________
Xen-devel mailing list
<a class="moz-txt-link-abbreviated" href="mailto:Xen-devel@lists.xen.org">Xen-devel@lists.xen.org</a>
<a class="moz-txt-link-freetext" href="http://lists.xen.org/xen-devel">http://lists.xen.org/xen-devel</a>
</pre>
    </blockquote>
    <br>
  </body>
</html>

--------------000708000409090103090702--


--===============0118481568339499055==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

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

--===============0118481568339499055==--