Merge pull request #1123 from NVIDIA/rel-24.9-cpicks

[release-24.9] cherrypick H200 NVL MIG and container-toolkit 1.17.2 changes

Author: tariq1890

assets/state-mig-manager/0400_configmap.yaml

@@ -201,7 +201,7 @@ data:
           mig-devices:
             "7g.96gb": 1
 
-      # GH200 144G HBM3e, H200-141GB, H100-96GB, GH200, H100 NVL, H800 NVL, H100-80GB, H800-80GB, A800-40GB, A800-80GB, A100-40GB, A100-80GB, A30-24GB, PG506-96GB
+      # GH200 144G HBM3e, H200-141GB, H200 NVL, H100-96GB, GH200, H100 NVL, H800 NVL, H100-80GB, H800-80GB, A800-40GB, A800-80GB, A100-40GB, A100-80GB, A30-24GB, PG506-96GB
       all-balanced:
         # GH200 144G HBM3e
         - device-filter: ["0x234810DE"]
@@ -212,8 +212,8 @@ data:
             "2g.36gb": 1
             "3g.72gb": 1
 
-        # H200 141GB
-        - device-filter: ["0x233510DE"]
+        # H200 141GB, H200 NVL
+        - device-filter: ["0x233510DE", "0x233B10DE"]
           devices: all
           mig-enabled: true
           mig-devices:

bundle/manifests/gpu-operator-certified.clusterserviceversion.yaml

@@ -207,7 +207,7 @@ spec:
     - name: dcgm-image
       image: nvcr.io/nvidia/cloud-native/dcgm@sha256:7b0ebd6c40a11b6484dc4385605372511e4e93132a44d2a3d6ec2e36c24e6783
     - name: container-toolkit-image
-      image: nvcr.io/nvidia/k8s/container-toolkit@sha256:bb6b55a5bd6419df3ca2d8ec0738b87491fc45e15587e613663890dc3a8e6e13
+      image: nvcr.io/nvidia/k8s/container-toolkit@sha256:f9d5652cba9ff27f9c4d17f5d09b4a5e4aa631b5cabb9cedf4d3e8c43d6847a4
     - name: driver-image
       image: nvcr.io/nvidia/driver@sha256:8a9a9e9470f64d340a7f3106a03e9622fa98e25368fbfb7ce9c416ad98f6d951
     - name: driver-image-535
@@ -862,7 +862,7 @@ spec:
                   - name: "GFD_IMAGE"
                     value: "nvcr.io/nvidia/k8s-device-plugin@sha256:7089559ce6153018806857f5049085bae15b3bf6f1c8bd19d8b12f707d087dea"
                   - name: "CONTAINER_TOOLKIT_IMAGE"
-                    value: "nvcr.io/nvidia/k8s/container-toolkit@sha256:bb6b55a5bd6419df3ca2d8ec0738b87491fc45e15587e613663890dc3a8e6e13"
+                    value: "nvcr.io/nvidia/k8s/container-toolkit@sha256:f9d5652cba9ff27f9c4d17f5d09b4a5e4aa631b5cabb9cedf4d3e8c43d6847a4"
                   - name: "DCGM_IMAGE"
                     value: "nvcr.io/nvidia/cloud-native/dcgm@sha256:7b0ebd6c40a11b6484dc4385605372511e4e93132a44d2a3d6ec2e36c24e6783"
                   - name: "DCGM_EXPORTER_IMAGE"

deployments/gpu-operator/values.yaml

@@ -231,7 +231,7 @@ toolkit:
   enabled: true
   repository: nvcr.io/nvidia/k8s
   image: container-toolkit
-  version: v1.17.0-ubuntu20.04
+  version: v1.17.2-ubuntu20.04
   imagePullPolicy: IfNotPresent
   imagePullSecrets: []
   env: []

go.mod

@@ -9,7 +9,7 @@ require (
 	github.com/NVIDIA/go-nvlib v0.7.0
 	github.com/NVIDIA/k8s-kata-manager v0.2.2
 	github.com/NVIDIA/k8s-operator-libs v0.0.0-20240826221728-249ba446fa35
-	github.com/NVIDIA/nvidia-container-toolkit v1.17.0
+	github.com/NVIDIA/nvidia-container-toolkit v1.17.2
 	github.com/davecgh/go-spew v1.1.2-0.20180830191138-d8f796af33cc
 	github.com/go-logr/logr v1.4.2
 	github.com/mittwald/go-helm-client v0.12.14

go.sum

@@ -30,8 +30,8 @@ github.com/NVIDIA/k8s-kata-manager v0.2.2 h1:+xVIp4yLfCjZ31Dfrm9LOKo4T47b4g+DV6X
 github.com/NVIDIA/k8s-kata-manager v0.2.2/go.mod h1:UGjGQUcpXTegwyOc5IwcyLTzPKwO9lOIkqw/qUzk8Q0=
 github.com/NVIDIA/k8s-operator-libs v0.0.0-20240826221728-249ba446fa35 h1:w9DXPTJCq9k2PVpdBQJrWE4vAmZcFaSHKLpM/xos9WI=
 github.com/NVIDIA/k8s-operator-libs v0.0.0-20240826221728-249ba446fa35/go.mod h1:sw6XRI5wq0Q+nSgaWa1Pyo/ZKxQebc70x6VIznDAxtM=
-github.com/NVIDIA/nvidia-container-toolkit v1.17.0 h1:L1kRJmoV5o42Ehnuh9lg959UxviTh2g2aneBOxjN7u0=
-github.com/NVIDIA/nvidia-container-toolkit v1.17.0/go.mod h1:R6bNf6ca0IjjACa0ncKGvsrx6zSjsgz8QkFyBDk5szU=
+github.com/NVIDIA/nvidia-container-toolkit v1.17.2 h1:iE6PK9SQH3HyDrOolu27xn3CJgURR3bDtnbfFrxdML8=
+github.com/NVIDIA/nvidia-container-toolkit v1.17.2/go.mod h1:R6bNf6ca0IjjACa0ncKGvsrx6zSjsgz8QkFyBDk5szU=
 github.com/Shopify/logrus-bugsnag v0.0.0-20171204204709-577dee27f20d h1:UrqY+r/OJnIp5u0s1SbQ8dVfLCZJsnvazdBP5hS4iRs=
 github.com/Shopify/logrus-bugsnag v0.0.0-20171204204709-577dee27f20d/go.mod h1:HI8ITrYtUY+O+ZhtlqUnD8+KwNPOyugEhfP9fdUIaEQ=
 github.com/alecthomas/template v0.0.0-20160405071501-a0175ee3bccc/go.mod h1:LOuyumcjzFXgccqObfd/Ljyb9UuFJ6TxHnclSeseNhc=

vendor/github.com/NVIDIA/nvidia-container-toolkit/internal/lookup/symlinks/symlink.go

@@ -45,7 +45,7 @@ github.com/NVIDIA/k8s-kata-manager/api/v1alpha1/config
 github.com/NVIDIA/k8s-operator-libs/api/upgrade/v1alpha1
 github.com/NVIDIA/k8s-operator-libs/pkg/consts
 github.com/NVIDIA/k8s-operator-libs/pkg/upgrade
-# github.com/NVIDIA/nvidia-container-toolkit v1.17.0
+# github.com/NVIDIA/nvidia-container-toolkit v1.17.2
 ## explicit; go 1.20
 github.com/NVIDIA/nvidia-container-toolkit/cmd/nvidia-ctk/system/create-dev-char-symlinks
 github.com/NVIDIA/nvidia-container-toolkit/internal/info/proc/devices