7:10 a.m.
Signed-off-by: Erik Skultety <eskultet(a)redhat.com>
---
docs/drvnodedev.html.in | 168 +++++++++++++++++++++++++++++++++++++++++++++++-
tools/virsh.pod | 7 +-
2 files changed, 171 insertions(+), 4 deletions(-)
diff --git a/docs/drvnodedev.html.in b/docs/drvnodedev.html.in
index 0a3870343..26c52dd0d 100644
--- a/docs/drvnodedev.html.in
+++ b/docs/drvnodedev.html.in
@@ -9,7 +9,7 @@
(historically also referred to as node devices) like USB, PCI, SCSI, and
network devices. This also includes various virtualization capabilities
which the aforementioned devices provide for utilization, for example
- SR-IOV, NPIV, DRM, etc.
+ SR-IOV, NPIV, MDEV, DRM, etc.
</p>
<p>
@@ -75,6 +75,7 @@
<code>storage</code> (<span class="since">Since
1.0.4</span>),
<code>scsi_generic</code> (<span
class="since">Since 1.0.7</span>),
<code>drm</code> (<span class="since">Since
3.1.0</span>), and
+ <code>mdev</code> (<span class="since">Since
3.4.0</span>).
This element can be nested in which case it further specifies a
device's capability. Refer to specific device types to see more values
for the <code>type</code> attribute which are exclusive.
@@ -185,5 +186,170 @@
...
<device></pre>
+ <h3><a name="MDEVCap">MDEV capability</a></h3>
+ <p>
+ A PCI device capable of creating mediated devices will include a nested
+ capability <code>mdev_types</code> which enumerates all supported mdev
+ types on the physical device, along with the type attributes available
+ through sysfs:
+ </p>
+
+ <dl>
+ <dt><code>type</code></dt>
+ <dd>
+ This element describes a mediated device type which acts as an
+ abstract template defining a resource allocation for instances of this
+ device type. The element has one attribute <code>id</code> which
holds
+ an official vendor-supplied identifier for the type.
+ <span class="since">Since 3.4.0</span>
+ </dd>
+
+ <dt><code>name</code></dt>
+ <dd>
+ The <code>name</code> element holds a vendor-supplied code name for
+ the given mediated device type. This is an optional element.
+ <span class="since">Since 3.4.0</span>
+ </dd>
+
+ <dt><code>deviceAPI</code></dt>
+ <dd>
+ The value of this element describes how an instance of the given type
+ will be presented to the guest by the VFIO framework.
+ <span class="since">Since 3.4.0</span>
+ </dd>
+
+ <dt><code>availableInstances</code></dt>
+ <dd>
+ This element reports the current state of resource allocation. In other
+ words, how many instances of the given type can still be successfully
+ created on the physical device.
+ <span class="since">Since 3.4.0</span>
+ </dd>
+ </dl>
+
+ <p>
+ For a more info about mediated devices, refer to the
+ <a href="#MDEV">paragraph below</a>.
+ </p>
+
+<pre>
+<device>
+...
+ <driver>
+ <name>nvidia</name>
+ </driver>
+ <capability type='pci'>
+...
+ <capability type='mdev_types'>
+ <type id='nvidia-11'>
+ <name>GRID M60-0B</name>
+ <deviceAPI>vfio-pci</deviceAPI>
+ <availableInstances>16</availableInstances>
+ </type>
+ <!-- Here would come the rest of the available mdev types -->
+ </capability>
+...
+ </capability>
+</device></pre>
+
+ <h2><a name="MDEV">Mediated devices
(MDEVs)</a></h2>
+ <p>
+ Mediated devices (<span class="since">Since 3.2.0</span>) are
software
+ devices defining resource allocation on the backing physical device which
+ in turn allows the parent physical device's resources to be divided into
+ several mediated devices, thus sharing the physical device's performance
+ among multiple guests. Unlike SR-IOV however, where a PCIe device appears
+ as multiple separate PCIe devices on the host's PCI bus, mediated devices
+ only appear on the mdev virtual bus. Therefore, no detach/reattach
+ procedure from/to the host driver procedure is involved even though
+ mediated devices are used in a direct device assignment manner.
+ </p>
+
+ <p>
+ The following sub-elements and attributes are exposed within the
+ <code>capability</code> element:
+ </p>
+
+ <dl>
+ <dt><code>type</code></dt>
+ <dd>
+ This element describes a mediated device type which acts as an
+ abstract template defining a resource allocation for instances of this
+ device type. The element has one attribute <code>id</code> which
holds
+ an official vendor-supplied identifier for the type.
+ <span class="since">Since 3.4.0</span>
+ </dd>
+
+ <dt><code>iommuGroup</code></dt>
+ <dd>
+ This element supports a single attribute <code>number</code> which
holds
+ the IOMMU group number the mediated device belongs to.
+ <span class="since">Since 3.4.0</span>
+ </dd>
+ </dl>
+
+ <h3>Example of a mediated device</h3>
+ <pre>
+<device>
+ <name>mdev_4b20d080_1b54_4048_85b3_a6a62d165c01</name>
+
<path>/sys/devices/pci0000:00/0000:00:02.0/4b20d080-1b54-4048-85b3-a6a62d165c01</path>
+ <parent>pci_0000_06_00_0</parent>
+ <driver>
+ <name>vfio_mdev</name>
+ </driver>
+ <capability type='mdev'>
+ <type id='nvidia-11'/>
+ <iommuGroup number='12'/>
+ <capability/>
+<device/></pre>
+
+ <p>
+ The support of mediated device's framework in libvirt's node device driver
+ covers the following features:
+ </p>
+
+ <ul>
+ <li>
+ list available mediated devices on the host
+ (<span class="since">Since 3.4.0</span>)
+ </li>
+ <li>
+ display device details
+ (<span class="since">Since 3.4.0</span>)
+ </li>
+ </ul>
+
+ <p>
+ Because mediated devices are instantiated from vendor specific templates,
+ simply called 'types', information describing these types is contained
+ within the parent device's capabilities
+ (see the example in <a href="#PCI">PCI host devices</a>).
+ </p>
+
+ <p>
+ To see the supported mediated device types on a specific physical device
+ use the following:
+ </p>
+
+ <pre>
+$ ls /sys/class/mdev_bus/<device>/mdev_supported_types</pre>
+
+ <p>
+ To manually instantiate a mediated device, use one of the following as a
+ reference:
+ </p>
+
+ <pre>
+$ uuidgen >
/sys/class/mdev_bus/<device>/mdev_supported_types/<type>/create
+...
+$ echo <UUID> >
/sys/class/mdev_bus/<device>/mdev_supported_types/<type>/create</pre>
+
+ <p>
+ Manual removal of a mediated device is then performed as follows:
+ </p>
+
+ <pre>
+$ echo 1 > /sys/bus/mdev/devices/<uuid>/remove</pre>
+
</body>
</html>
diff --git a/tools/virsh.pod b/tools/virsh.pod
index 727acf6e6..d69c8d87f 100644
--- a/tools/virsh.pod
+++ b/tools/virsh.pod
@@ -3177,10 +3177,11 @@ for HBA).
List all of the devices available on the node that are known by libvirt.
I<cap> is used to filter the list by capability types, the types must be
-separated by comma, e.g. --cap pci,scsi, valid capability types include
+separated by comma, e.g. --cap pci,scsi. Valid capability types include
'system', 'pci', 'usb_device', 'usb', 'net',
'scsi_host', 'scsi_target',
-'scsi', 'storage', 'fc_host', 'vports',
'scsi_generic', 'drm'.If I<--tree>
-is used, the output is formatted in a tree representing parents of each
+'scsi', 'storage', 'fc_host', 'vports',
'scsi_generic', 'drm', 'mdev',
+'mdev_types'.
+If I<--tree> is used, the output is formatted in a tree representing parents of
each
node. I<cap> and I<--tree> are mutually exclusive.
=item B<nodedev-reattach> I<nodedev>
--
2.13.0