Merge pull request #57 from archlitchi/dev-vgpu-1219

adjust docs ,submodule libvgpu, update ci
volcano-sh · Feb 4, 2024 · 504ec3e · 504ec3e
2 parents b90971f + 4002de6
commit 504ec3e
Show file tree

Hide file tree

Showing 83 changed files with 31 additions and 11,993 deletions.
diff --git a/.github/workflows/main.yml b/.github/workflows/main.yml
@@ -5,7 +5,9 @@ on:
  - v[0-9]+.[0-9]+.[0-9]+
  branches: [master,dev-vgpu-1219]
  pull_request:
- branches: [master,dev-vgpu-1219]
+ branches: [master]
+ pull_request_target:
+ branches: [dev-vgpu-1219]
 
 jobs:
  build:
@@ -15,6 +17,12 @@ jobs:
  - uses: actions/setup-go@v2
  with:
  go-version: "^1.19.x"
+
+ - name: Checkout submodule
+ uses: Mushus/checkout-submodule@v1.0.1
+ with:
+ submodulePath: libvgpu 
+
  - run: go version
  # Runs a set of commands to initialize and analyze with FOSSA
 

diff --git a/.gitmodules b/.gitmodules
@@ -0,0 +1,3 @@
+[submodule "libvgpu"]
+ path = libvgpu
+ url = https://github.com/Project-HAMi/HAMi-core.git
diff --git a/README.md b/README.md
@@ -1,7 +1,8 @@
 # Volcano device plugin for Kubernetes
 
 **Note**:
-This is based on [Nvidia Device Plugin](https://github.com/NVIDIA/k8s-device-plugin) to support soft isolation of GPU card.
+This is based on [Nvidia Device Plugin](https://github.com/NVIDIA/k8s-device-plugin), it uses [HAMi-core](https://github.com/Project-HAMi/HAMi-core) to support hard isolation of GPU card.
+
 And collaborate with volcano, it is possible to enable GPU sharing.
 
 ## Table of Contents
@@ -23,6 +24,7 @@ The Volcano device plugin for Kubernetes is a Daemonset that allows you to autom
 - Expose the number of GPUs on each node of your cluster
 - Keep track of the health of your GPUs
 - Run GPU enabled containers in your Kubernetes cluster.
+- Enforce hard resource limit in container.
 
 This repository contains Volcano's official implementation of the [Kubernetes device plugin](https://github.com/kubernetes/community/blob/master/contributors/design-proposals/resource-management/device-plugin.md).
 

diff --git a/docker/amd64/Dockerfile.vgpu-ubuntu20.04 b/docker/amd64/Dockerfile.vgpu-ubuntu20.04
@@ -35,7 +35,7 @@ RUN export CGO_LDFLAGS_ALLOW='-Wl,--unresolved-symbols=ignore-in-object-files' &
  go build -ldflags="-s -w" -o volcano-vgpu-device-plugin ./cmd/vgpu
 
 FROM $NVIDIA_IMAGE AS NVBUILD
-COPY ./lib/nvidia/libvgpu /libvgpu
+COPY ./libvgpu /libvgpu
 WORKDIR /libvgpu
 RUN apt-get -y update; apt-get -y install wget
 RUN wget https://cmake.org/files/v3.19/cmake-3.19.8-Linux-x86_64.tar.gz

diff --git a/lib/nvidia/libvgpu/CMakeLists.txt b/lib/nvidia/libvgpu/CMakeLists.txt
diff --git a/lib/nvidia/libvgpu/README.md b/lib/nvidia/libvgpu/README.md
diff --git a/lib/nvidia/libvgpu/build.sh b/lib/nvidia/libvgpu/build.sh
diff --git a/lib/nvidia/libvgpu/dockerfiles/Dockerfile.aarch64 b/lib/nvidia/libvgpu/dockerfiles/Dockerfile.aarch64
diff --git a/lib/nvidia/libvgpu/dockerfiles/Dockerfile.x86_64 b/lib/nvidia/libvgpu/dockerfiles/Dockerfile.x86_64
diff --git a/lib/nvidia/libvgpu/docs/images/sample_nvidia-smi.png b/lib/nvidia/libvgpu/docs/images/sample_nvidia-smi.png
diff --git a/lib/nvidia/libvgpu/framework_test/Dockerfile b/lib/nvidia/libvgpu/framework_test/Dockerfile
diff --git a/lib/nvidia/libvgpu/framework_test/README.md b/lib/nvidia/libvgpu/framework_test/README.md
diff --git a/lib/nvidia/libvgpu/framework_test/clients/dockerclient.go b/lib/nvidia/libvgpu/framework_test/clients/dockerclient.go
diff --git a/lib/nvidia/libvgpu/framework_test/clients/nvmlclient.go b/lib/nvidia/libvgpu/framework_test/clients/nvmlclient.go
diff --git a/lib/nvidia/libvgpu/framework_test/config.toml b/lib/nvidia/libvgpu/framework_test/config.toml