docs: document vdpa device type

Signed-off-by: Adrian Moreno <amorenoz@redhat.com>

README.md

@@ -181,6 +181,15 @@ This plugin creates device plugin endpoints based on the configurations given in
                 "isRdma": true
             }
         },
+        {
+            "resourceName": "ct6dx_vdpa_vhost",
+            "selectors": {
+                "vendors": ["15b3"],
+                "devices": ["101e"],
+                "drivers": ["mlx5_core"],
+                "vdpaType": "vhost"
+            }
+        },
         {
             "resourceName": "intel_fpga",
             "deviceType": "accelerator",
@@ -233,6 +242,7 @@ This selector is applicable when "deviceType" is "netDevice"(note: this is defau
 | "ddpProfiles" | N        | A map of device selectors                                      | `string` list Default: `null`                     | "ddpProfiles": ["GTPv1-C/U IPv4/IPv6 payload"]                                                   |
 | "isRdma"      | N        | Mount RDMA resources                                           | `bool`  values `true` or `false` Default: `false` | "isRdma": `true`                                                                                 |
 | "needVhostNet"| N        | Share /dev/vhost-net                                           | `bool`  values `true` or `false` Default: `false` | "needVhostNet": `true`                                                                           |
+| "vdpaType"    | N        | The type of vDPA device (virtio, vhost)                        | `string` values `vhost` or `virtio` Default: `null` | "vdpaType": "vhost"                                                                           |
 
 
 [//]: # (The tables above generated using: https://ozh.github.io/ascii-tables/)

docs/README.md

@@ -7,3 +7,4 @@ This page contains supplimentary documention that users may find useful for vari
 * [Running RDMA application in Kubernetes](rdma/)
 * [SR-IOV Network Device Plugin with DDP](ddp/)
 * [Using node specific config file for running device plugin DaemonSet](config-file)
+* [Using vDPA devices in Kuberenets](vdpa/)

docs/vdpa/README.md

@@ -0,0 +1,68 @@
+# Using vDPA devices in Kubernetes
+## Introduction to vDPA
+vDPA (Virtio DataPath Acceleration) is a technology that enables the acceleration
+of virtIO devices while allowing the implementations of such devices
+(e.g: NIC vendors) to use their own control plane.
+
+The consumers of the virtIO devices (VMs or containers) interact with the devices
+using the standard virtIO datapath and virtio-compatible control paths (virtIO, vhost).
+While the data-plane is mapped directly to the accelerator device, the contol-plane
+gets translated the vDPA kernel framework.
+
+The vDPA kernel framework is composed of a vdpa bus (/sys/bus/vdpa), vdpa devices
+(/sys/bus/vdpa/devices) and vdpa drivers (/sys/bus/vdpa/drivers).
+Currently, two vdpa drivers are implemented:
+*  virtio_vdpa: Exposes the device as a virtio-net netdev
+*  vhost_vdpa: Exposes the device as a vhost-vdpa device. This device uses an extension
+of the vhost-net protocol to allow userspace applications access the rings directly
+
+For more information about the vDPA framework, read the article on
+[LWN.net](https://lwn.net/Articles/816063/) or the blog series written by one of the
+main authors ([part 1](https://www.redhat.com/en/blog/vdpa-kernel-framework-part-1-vdpa-bus-abstracting-hardware),
+[part 2](https://www.redhat.com/en/blog/vdpa-kernel-framework-part-2-vdpa-bus-drivers-kernel-subsystem-interactions),
+[part3](https://www.redhat.com/en/blog/vdpa-kernel-framework-part-3-usage-vms-and-containers))
+
+## vDPA Management
+Currently, the management of vDPA devices is performed using the sysfs interface exposed
+by the vDPA Framework. However, in order to decouple the management of vdpa devices from
+the SR-IOV Device Plugin functionality, this low-level management is done in an external
+library called [go-vdpa](https://github.com/redhat-virtio-net/govdpa).
+
+In the context of the SR-IOV Device Plugin and the SR-IOV CNI, the current plan is to
+support only 1:1 mappings between SR-IOV VFs and vDPA devices despite the fact that
+the vDPA Framework might support 1:N mappings.
+
+## Tested NICs:
+* Mellanox ConnectX®-6 DX *
+
+## Prerequisites
+* Linux Kernel >= 5.12
+* iproute >= 5.14
+
+## vDPA device creation
+Insert the vdpa kernel modules if not present:
+
+    $ modprobe vdpa
+    $ modprobe virtio-vdpa
+    $ modprobe vhost-vdpa
+
+Create vdpa using the vdpa management tool integrated into iproute2, e.g:
+
+    $ vdpa mgmtdev show
+    pci/0000:65:00.2:
+      supported_classes net
+    $ vdpa dev add name vdpa2 mgmtdev pci/0000:65:00.2
+    $ vdpa dev list
+    vdpa2: type network mgmtdev pci/0000:65:00.2 vendor_id 5555 max_vqs 16 max_vq_size 256
+
+## Bind the desired vdpa driver
+The vdpa bus works similar to the pci bus. To unbind a driver from a device, run:
+
+    echo ${DEV_NAME} > /sys/bus/vdpa/devices/${DEV_NAME}/driver/unbind
+
+To bind a driver to a device, run:
+
+    echo ${DEV_NAME} > /sys/bus/vdpa/drivers/${DRIVER_NAME}/bind
+
+## Configure the SR-IOV Device Plugin
+See the sample [configMap](configMap.yaml) for an example of how to configure a vDPA device

docs/vdpa/configMap.yaml

@@ -0,0 +1,30 @@
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: sriovdp-config
+  namespace: kube-system
+data:
+  config.json: |
+    {
+        "resourceList": [{
+            {
+                "resourceName": "vdpa_mlx_virtio",
+                "selectors": {
+                    "vendors": ["15b3"],
+                    "devices": ["101e"],
+                    "drivers": ["mlx5_core"],
+                    "vdpaType": "virtio"
+                }
+            },
+            {
+                "resourceName": "vdpa_mlx_vhost",
+                "selectors": {
+                    "vendors": ["15b3"],
+                    "devices": ["101e"],
+                    "drivers": ["mlx5_core"],
+                    "vdpaType": "vhost"
+                }
+            }
+        ]
+    }
+