[libvirt] [PATCHv2 3/6] conf: add <driver intremap> to <iommu>
Ján Tomko
jtomko at redhat.com
Wed Apr 26 07:54:23 UTC 2017
On Mon, Apr 24, 2017 at 05:42:07PM -0400, John Ferlan wrote:
>
>
>On 04/20/2017 08:19 AM, Ján Tomko wrote:
>> Add a new attribute to control interrupt remapping.
>>
>> https://bugzilla.redhat.com/show_bug.cgi?id=1427005
>> ---
>> docs/formatdomain.html.in | 22 ++++++++++++-
>> docs/schemas/domaincommon.rng | 9 +++++
>> src/conf/domain_conf.c | 38 +++++++++++++++++++---
>> src/conf/domain_conf.h | 1 +
>> .../qemuxml2argv-intel-iommu-irqchip.xml | 4 ++-
>> 5 files changed, 68 insertions(+), 6 deletions(-)
>>
>> diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in
>> index abf089a..f5a8e76 100644
>> --- a/docs/formatdomain.html.in
>> +++ b/docs/formatdomain.html.in
>> @@ -7335,7 +7335,9 @@ qemu-kvm -net nic,model=? /dev/null
>> <pre>
>> ...
>> <devices>
>> - <iommu model='intel'/>
>> + <iommu model='intel'>
>> + <driver intremap='on'/>
>> + </iommu>
>> </devices>
>> ...
>> </pre>
>> @@ -7346,6 +7348,24 @@ qemu-kvm -net nic,model=? /dev/null
>> Currently only the <code>intel</code> model is supported.
>> </p>
>> </dd>
>> + <dt><code>driver</code></dt>
>> + <dd>
>> + <p>
>> + The <code>driver</code> subelement can be used to configure
>> + additional options:
>> + </p>
>> + <dl>
>> + <dt><code>intremap</code></dt>
>> + <dd>
>> + <p>
>> + The <code>intremap</code> attribute with possible values
>> + <code>on</code> and <code>off</code> can be used to
>> + turn on interrupt remapping. <span class="since">Since 3.3.0</span>
>> + (QEMU only)
>
>It seems there is a relationship between this parameter and irqchip
>mode? IOW: Is it a requirement that irqchip be "split" or "on"? It's
>difficult to ascertain from the bz, but if so, then I'd expect a domain
>conf post processing in this patch.
>
Per http://wiki.qemu.org/Features/VT-d:
Currently, interrupt remapping does not support full kernel irqchip,
only "split" and "off" are supported.
We should not put current QEMU implementation limits into XML post
processing.
>Reading the bz it seems though that this parameter is optional for
>"certain" devices; however, for "general emulated devices" it's what is
>used to enable VT-d protection, but that could also be read as if
>'intel-iommu' is enabled, then VT-d protection is on by default with
>'intremap' just having some other affect.
Interrupt remapping is part of the VT-d feature.
>
>Is there something we could describe here at a very high level to
>describe the usefulness/need for this parameter?
>
IMO that is out of scope of the element description.
Jan
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: Digital signature
URL: <http://listman.redhat.com/archives/libvir-list/attachments/20170426/913769a8/attachment-0001.sig>
More information about the libvir-list
mailing list