[intel-gpu-plugin] Allow GPU sharing (#658)

charts/intel-gpu-plugin/Chart.yaml

@@ -1,7 +1,7 @@
 apiVersion: v2
 name: intel-gpu-plugin
-version: 1.0.1
-appVersion: 0.18.1
+version: 1.1.0
+appVersion: 0.19.0
 description: The Intel GPU plugin facilitates offloading the processing of computation intensive workloads to GPU hardware
 keywords:
   - kubernetes

charts/intel-gpu-plugin/README.md

@@ -1,6 +1,6 @@
 # intel-gpu-plugin
 
-![Version: 1.0.1](https://img.shields.io/badge/Version-1.0.1-informational?style=flat-square) ![AppVersion: 0.18.1](https://img.shields.io/badge/AppVersion-0.18.1-informational?style=flat-square)
+![Version: 1.1.0](https://img.shields.io/badge/Version-1.1.0-informational?style=flat-square) ![AppVersion: 0.19.0](https://img.shields.io/badge/AppVersion-0.19.0-informational?style=flat-square)
 
 The Intel GPU plugin facilitates offloading the processing of computation intensive workloads to GPU hardware
 
@@ -101,6 +101,7 @@ affinity:
 | serviceAccount.annotations | object | `{}` |  |
 | serviceAccount.create | bool | `true` |  |
 | serviceAccount.name | string | `""` |  |
+| shared_device_number | int | 1 | Number of containers sharing the same GPU device  |
 | strategyType | string | `"Recreate"` | Specifies the strategy used to replace old Pods by new ones |
 | tolerations | list | `[]` |  |
 
@@ -110,7 +111,21 @@ All notable changes to this application Helm chart will be documented in this fi
 
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/), and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
-### [0.18.1]
+### [1.1.0]
+
+#### Added
+
+- This chart now supports the `shared-dev-num` value. This allows setting the number of containers sharing the same GPU device.
+
+#### Changed
+
+- Bumped the image version to 0.19.0.
+
+#### Fixed
+
+- Changelog versions
+
+### [1.0.1]
 
 #### Added
 
@@ -124,7 +139,8 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 
 - N/A
 
-[0.18.1]: #0.18.1
+[1.1.0]: #1.1.0
+[1.0.1]: #1.0.1
 
 ## Support
 
@@ -134,4 +150,4 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 - Join our [Discord](https://discord.gg/sTMX7Vh) community
 
 ----------------------------------------------
-Autogenerated from chart metadata using [helm-docs v1.5.0](https://github.com/norwoodj/helm-docs/releases/v1.5.0)
+Autogenerated from chart metadata using [helm-docs v1.5.0](https://github.com/norwoodj/helm-docs/releases/v1.5.0)

charts/intel-gpu-plugin/README_CHANGELOG.md.gotmpl

@@ -9,7 +9,22 @@ All notable changes to this application Helm chart will be documented in this fi
 
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/), and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
-### [0.18.1]
+### [1.1.0]
+
+#### Added
+
+- This chart now supports the `shared-dev-num` value. This allows setting the number of containers sharing the same GPU device.
+
+#### Changed
+
+- Bumped the image version to 0.19.0.
+
+#### Fixed
+
+- Changelog versions
+
+
+### [1.0.1]
 
 #### Added
 
@@ -23,5 +38,6 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 
 - N/A
 
-[0.18.1]: #0.18.1
+[1.1.0]: #1.1.0
+[1.0.1]: #1.0.1
 {{- end -}}

charts/intel-gpu-plugin/templates/daemonset.yaml

@@ -36,6 +36,9 @@ spec:
               valueFrom:
                 fieldRef:
                   fieldPath: spec.nodeName
+          args:
+            - -shared-dev-num
+            - {{ .Values.shared_device_number | quote }}
           volumeMounts:
             - name: devfs
               mountPath: /dev/dri

charts/intel-gpu-plugin/values.yaml

@@ -6,7 +6,7 @@ image:
   # -- Image pull policy
   pullPolicy: IfNotPresent
   # -- Image tag. Possible values listed [here](https://hub.docker.com/r/intel/intel-gpu-plugin/tags)
-  tag: 0.18.1
+  tag: 0.19.0
 
 # -- Specifies the strategy used to replace old Pods by new ones
 strategyType: Recreate
@@ -24,6 +24,10 @@ serviceAccount:
   # If not set and create is true, a name is generated using the fullname template
   name: ""
 
+# -- Number of containers sharing the same GPU device
+# @default -- 1
+shared_device_number: 1
+
 # -- Key-value pairs to add as pod annotations
 podAnnotations: {}