Merge pull request #1154 from NVIDIA/cherry-picks-for-24.9.1

Cherry picks for 24.9.1

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

@@ -18,7 +18,7 @@ metadata:
     features.operators.openshift.io/cnf: "false"
     features.operators.openshift.io/cni: "false"
     features.operators.openshift.io/csi: "false"
-    olm.skipRange: '>=1.9.0 <24.9.0'
+    olm.skipRange: '>=1.9.0 <24.9.1'
     alm-examples: |-
       [
         {
@@ -195,7 +195,7 @@ metadata:
     provider: NVIDIA
     repository: http://github.com/NVIDIA/gpu-operator
     support: NVIDIA
-  name: gpu-operator-certified.v24.9.0
+  name: gpu-operator-certified.v24.9.1
   namespace: placeholder
 spec:
   apiservicedefinitions: {}
@@ -207,7 +207,7 @@ spec:
     - name: dcgm-image
       image: nvcr.io/nvidia/cloud-native/dcgm@sha256:e8398b2451996e5c64fc3855e7a3c93b667a85f69c0e259049d3b0012c21545d
     - name: container-toolkit-image
-      image: nvcr.io/nvidia/k8s/container-toolkit@sha256:f9d5652cba9ff27f9c4d17f5d09b4a5e4aa631b5cabb9cedf4d3e8c43d6847a4
+      image: nvcr.io/nvidia/k8s/container-toolkit@sha256:9668652f52eb47e3c6791da9c4293839a21e925ecee37161bda90f3b61bae290
     - name: driver-image
       image: nvcr.io/nvidia/driver@sha256:78e00fc8aa72f5f4925d54b4f9fbb725ca2168e890d90eadd6a497b7f31ccc18
     - name: driver-image-535
@@ -223,13 +223,13 @@ spec:
     - name: mig-manager-image
       image: nvcr.io/nvidia/cloud-native/k8s-mig-manager@sha256:40830d3561c14743f484d45b498141f9e86b1308e16fae3978110783927264ab
     - name: init-container-image
-      image: nvcr.io/nvidia/cuda@sha256:748a2c5178e5c5811b66183bd0ce87d9fdccf992c0ad9b1a5076841e45533190
+      image: nvcr.io/nvidia/cuda@sha256:b3b340847db9abd0c39703278f5638d3e65e1369e9a526a19b3138d5a6563527
     - name: gpu-operator-validator-image
       image: registry.gitlab.com/nvidia/kubernetes/gpu-operator/staging/gpu-operator-validator:main-latest
     - name: k8s-driver-manager-image
       image: nvcr.io/nvidia/cloud-native/k8s-driver-manager@sha256:b072c5793be65eee556eaff1b9cbbd115a1ef29982be95b2959adfcb4bc72382
     - name: vfio-manager-image
-      image: nvcr.io/nvidia/cuda@sha256:748a2c5178e5c5811b66183bd0ce87d9fdccf992c0ad9b1a5076841e45533190
+      image: nvcr.io/nvidia/cuda@sha256:b3b340847db9abd0c39703278f5638d3e65e1369e9a526a19b3138d5a6563527
     - name: sandbox-device-plugin-image
       image: nvcr.io/nvidia/kubevirt-gpu-device-plugin@sha256:4ffa1cd2a6497eb647a89ed259dcfb007554737b9d80f69bc173a2c3cd72a1da
     - name: vgpu-device-manager-image
@@ -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:f9d5652cba9ff27f9c4d17f5d09b4a5e4aa631b5cabb9cedf4d3e8c43d6847a4"
+                    value: "nvcr.io/nvidia/k8s/container-toolkit@sha256:9668652f52eb47e3c6791da9c4293839a21e925ecee37161bda90f3b61bae290"
                   - name: "DCGM_IMAGE"
                     value: "nvcr.io/nvidia/cloud-native/dcgm@sha256:e8398b2451996e5c64fc3855e7a3c93b667a85f69c0e259049d3b0012c21545d"
                   - name: "DCGM_EXPORTER_IMAGE"
@@ -882,9 +882,9 @@ spec:
                   - name: "MIG_MANAGER_IMAGE"
                     value: "nvcr.io/nvidia/cloud-native/k8s-mig-manager@sha256:40830d3561c14743f484d45b498141f9e86b1308e16fae3978110783927264ab"
                   - name: "CUDA_BASE_IMAGE"
-                    value: "nvcr.io/nvidia/cuda@sha256:748a2c5178e5c5811b66183bd0ce87d9fdccf992c0ad9b1a5076841e45533190"
+                    value: "nvcr.io/nvidia/cuda@sha256:b3b340847db9abd0c39703278f5638d3e65e1369e9a526a19b3138d5a6563527"
                   - name: "VFIO_MANAGER_IMAGE"
-                    value: "nvcr.io/nvidia/cuda@sha256:748a2c5178e5c5811b66183bd0ce87d9fdccf992c0ad9b1a5076841e45533190"
+                    value: "nvcr.io/nvidia/cuda@sha256:b3b340847db9abd0c39703278f5638d3e65e1369e9a526a19b3138d5a6563527"
                   - name: "SANDBOX_DEVICE_PLUGIN_IMAGE"
                     value: "nvcr.io/nvidia/kubevirt-gpu-device-plugin@sha256:4ffa1cd2a6497eb647a89ed259dcfb007554737b9d80f69bc173a2c3cd72a1da"
                   - name: "VGPU_DEVICE_MANAGER_IMAGE"
@@ -921,5 +921,5 @@ spec:
   maturity: stable
   provider:
     name: NVIDIA Corporation
-  version: 24.9.0
-  replaces: gpu-operator-certified.v24.6.2
+  version: 24.9.1
+  replaces: gpu-operator-certified.v24.9.0

deployments/gpu-operator/values.yaml

@@ -83,7 +83,7 @@ operator:
   initContainer:
     image: cuda
     repository: nvcr.io/nvidia
-    version: 12.6.2-base-ubi9
+    version: 12.6.3-base-ubi9
     imagePullPolicy: IfNotPresent
   tolerations:
   - key: "node-role.kubernetes.io/master"
@@ -231,7 +231,7 @@ toolkit:
   enabled: true
   repository: nvcr.io/nvidia/k8s
   image: container-toolkit
-  version: v1.17.2-ubuntu20.04
+  version: v1.17.3-ubuntu20.04
   imagePullPolicy: IfNotPresent
   imagePullSecrets: []
   env: []
@@ -487,7 +487,7 @@ vfioManager:
   enabled: true
   repository: nvcr.io/nvidia
   image: cuda
-  version: 12.6.2-base-ubi9
+  version: 12.6.3-base-ubi9
   imagePullPolicy: IfNotPresent
   imagePullSecrets: []
   env: []

docker/Dockerfile

@@ -14,7 +14,7 @@
 
 ARG GOLANG_VERSION=x.x.x
 
-FROM nvcr.io/nvidia/cuda:12.6.2-base-ubi9 as builder
+FROM nvcr.io/nvidia/cuda:12.6.3-base-ubi9 as builder
 
 RUN dnf install -y wget make git gcc
 
@@ -55,7 +55,7 @@ ARG VERSION="unknown"
 ARG GIT_COMMIT="unknown"
 RUN make gpu-operator
 
-FROM nvcr.io/nvidia/cuda:12.6.2-base-ubi9
+FROM nvcr.io/nvidia/cuda:12.6.3-base-ubi9
 
 # Remove CUDA libs(compat etc) in favor of libs installed by the NVIDIA driver
 RUN dnf remove -y cuda-*

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 v1.17.2
+	github.com/NVIDIA/nvidia-container-toolkit v1.17.3
 	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 v1.17.2 h1:iE6PK9SQH3HyDrOolu27xn3CJgURR3bDtnbfFrxdML8=
-github.com/NVIDIA/nvidia-container-toolkit v1.17.2/go.mod h1:R6bNf6ca0IjjACa0ncKGvsrx6zSjsgz8QkFyBDk5szU=
+github.com/NVIDIA/nvidia-container-toolkit v1.17.3 h1:vCJ9IY/YNcwNiMv7anAWEmUW8Xocqdmv73V98MXTrxo=
+github.com/NVIDIA/nvidia-container-toolkit v1.17.3/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=

validator/Dockerfile

@@ -15,7 +15,7 @@
 ARG CUDA_SAMPLE_IMAGE=undefined
 ARG GOLANG_VERSION=x.x.x
 
-FROM nvcr.io/nvidia/cuda:12.6.2-base-ubi9 as build
+FROM nvcr.io/nvidia/cuda:12.6.3-base-ubi9 as build
 
 RUN dnf install -y wget make git gcc
 
@@ -51,7 +51,7 @@ FROM ${CUDA_SAMPLE_IMAGE} AS sample-builder
 RUN mkdir /artifacts
 RUN cp /cuda-samples/vectorAdd /artifacts/vectorAdd
 
-FROM nvcr.io/nvidia/cuda:12.6.2-base-ubi9
+FROM nvcr.io/nvidia/cuda:12.6.3-base-ubi9
 
 # Remove CUDA libs(compat etc) in favor of libs installed by the NVIDIA driver
 RUN dnf remove -y cuda-*

vendor/modules.txt

@@ -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 v1.17.2
+# github.com/NVIDIA/nvidia-container-toolkit v1.17.3
 ## 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

versions.mk

@@ -17,7 +17,7 @@
 # To re-generate a bundle for another specific version without changing the standard setup, you can:
 # - use the VERSION as arg of the bundle target (e.g make bundle VERSION=0.0.2)
 # - use environment variables to overwrite this value (e.g export VERSION=0.0.2)
-VERSION ?= v24.9.0
+VERSION ?= v24.9.1
 
 GOLANG_VERSION ?= 1.22.8