diff --git a/docs/drivers/installation.md b/docs/drivers/installation.md index e54ced07b..c97eb8d8b 100644 --- a/docs/drivers/installation.md +++ b/docs/drivers/installation.md @@ -281,7 +281,7 @@ To check the full spec of `DeviceConfig` definition run `kubectl get crds device |---------------------------------------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------| | `enable` | Set to true for installing out-of-tree driver.
Set to false to skip driver install and use inbox/pre-installed driver. | `true` | | `blacklist` | Set to true to have the operator init the node labeller DaemonSet and add `amdgpu` to the selected worker nodes' modprobe blacklist.
Set to false to remove `amdgpu` from the selected nodes' modprobe blacklist. | `false` | -| `version` | ROCm driver version (e.g., "6.2.2").
See ROCm Versions: https://rocm.docs.amd.com/en/latest/release/versions.html | Ubuntu: `6.1.3`
CoreOS: `6.2.2` | +| `version` | amdgpu driver version (e.g., "6.2.2").
See amdgpu Versions: https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/release/versions.html | Ubuntu: `6.1.3`
CoreOS: `6.2.2` | | `image` | Registry URL and repository (without tag).
Note: Operator manages tags automatically. | Vanilla k8s: `image-registry:5000/$MOD_NAMESPACE/amdgpu_kmod`
OpenShift: `image-registry.openshift-image-registry.svc:5000/$MOD_NAMESPACE/amdgpu_kmod` | | `imageRegistrySecret.name` | Name of registry credentials secret to pull/push driver image. | | | `imageRegistryTLS.insecure` | If true, check if the container image already exists using plain HTTP. | `false` | diff --git a/docs/drivers/precompiled-driver.md b/docs/drivers/precompiled-driver.md index 1d3c71a45..0bcb322a7 100644 --- a/docs/drivers/precompiled-driver.md +++ b/docs/drivers/precompiled-driver.md @@ -15,7 +15,7 @@ Users could prepare pre-compiled driver images in advance and import them into t KMM determines the appropriate driver image based on the combination of: 1. Worker node OS information -2. Requested ROCm driver version +2. Requested amdgpu driver version ### Image Tag Format diff --git a/docs/drivers/upgrading.md b/docs/drivers/upgrading.md index c1645ca54..471894142 100644 --- a/docs/drivers/upgrading.md +++ b/docs/drivers/upgrading.md @@ -70,7 +70,7 @@ To check the full spec of upgrade configuration run kubectl get crds deviceconfi | `maxUnavailableNodes` | Maximum number (or Percentage) of nodes which can be unavailable (cordoned) in the cluster | `25%` | | `rebootRequired` | Reboot the node after driver upgrade is done. Waits for 60 mins post reboot before declaring as failed | `true` | -**Warning**: When using ROCm drivers version 6.3 and below, a known issue may prevent the driver upgrade from fully completing unless the node is rebooted. As a workaround, we strongly recommend setting the `rebootRequired` field to `true` in your upgrade policy. This ensures that a reboot is triggered after the driver upgrade, allowing the new driver to be fully loaded. This workaround should be applied until a permanent fix is provided in a future release. +**Warning**: When using amdgpu driver versions 6.3 and below, a known issue may prevent the driver upgrade from fully completing unless the node is rebooted. As a workaround, we strongly recommend setting the `rebootRequired` field to `true` in your upgrade policy. This ensures that a reboot is triggered after the driver upgrade, allowing the new driver to be fully loaded. This workaround should be applied until a permanent fix is provided in a future release. #### `driver.upgradePolicy.nodeDrainPolicy` Parameters diff --git a/docs/knownlimitations.md b/docs/knownlimitations.md index baa99604f..169e8011c 100644 --- a/docs/knownlimitations.md +++ b/docs/knownlimitations.md @@ -99,11 +99,11 @@ - *Root Cause:* DCM does not currently support standalone installation via a Debian package like other standalone components of the GPU Operator. We will be adding a Debian package to support standalone bare metal installations in the next release of DCM. - *Recommendation:* Those wishing to use GPU partitioning in a bare metal environment should instead use the standalone docker image for DCM. Alternatively users can use amd-smi to change partitioning modes. See [amdgpu-docs documentation](https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/gpu-partitioning/mi300x/quick-start-guide.html) for how to do this. -16. **The GPU Operator will report an error when ROCm driver install version doesn't match the version string in the [Radeon Repo](https://repo.radeon.com/rocm/apt/).** +16. **The GPU Operator will report an error when amdgpu driver install version doesn't match the version string in the [Radeon Repo](https://repo.radeon.com/rocm/apt/).** - *Impact:* The DeviceConfig will report an error if you specify `"6.4.0"` or `"6.3.0"` for the `spec.driver.version`. - *Root Cause:* The version specified in the CR would still have to match the version string on Radeon repo. - - *Recommendation:* Although this will be fixed in a future version of the GPU Operator, for the time being you will instead need to specific `"6.4"` or `"6.3"` when installing those versions of the ROCm amdgpu driver. + - *Recommendation:* Although this will be fixed in a future version of the GPU Operator, for the time being you will instead need to specify `"6.4"` or `"6.3"` when installing those versions of the ROCm amdgpu driver. 17. **Device plugin crashes when unable to establish watcher on kubelet device plugin path** diff --git a/docs/releasenotes.md b/docs/releasenotes.md index 261449683..f40ea5299 100644 --- a/docs/releasenotes.md +++ b/docs/releasenotes.md @@ -346,11 +346,11 @@ The AMD GPU Operator v1.3.0 release introduces new features, most notably of whi - *Root Cause:* DCM does not currently support standalone installation via a Debian package like other standalone components of the GPU Operator. We will be adding a Debian package to support standalone bare metal installations in the next release of DCM. - *Recommendation:* Those wishing to use GPU partitioning in a bare metal environment should instead use the standalone docker image for DCM. Alternatively users can use amd-smi to change partitioning modes. See [amdgpu-docs documentation](https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/gpu-partitioning/mi300x/quick-start-guide.html) for how to do this. -2. **The GPU Operator will report an error when ROCm driver install version doesn't match the version string in the [Radeon Repo](https://repo.radeon.com/rocm/apt/).** +2. **The GPU Operator will report an error when amdgpu driver install version doesn't match the version string in the [Radeon Repo](https://repo.radeon.com/rocm/apt/).** - *Impact:* The DeviceConfig will report an error if you specify `"6.4.0"` or `"6.3.0"` for the `spec.driver.version`. - *Root Cause:* The version specified in the CR would still have to match the version string on Radeon repo. - - *Recommendation:* Although this will be fixed in a future version of the GPU Operator, for the time being you will instead need to specific `"6.4"` or `"6.3"` when installing those versions of the ROCm amdgpu driver. + - *Recommendation:* Although this will be fixed in a future version of the GPU Operator, for the time being you will instead need to specify `"6.4"` or `"6.3"` when installing those versions of the ROCm amdgpu driver. > **Note:** All current and historical limitations for the GPU Operator, including their latest statuses and any associated workarounds or fixes, are tracked in the following documentation page: [Known Issues and Limitations](https://instinct.docs.amd.com/projects/gpu-operator/en/latest/knownlimitations.html). Please refer to this page regularly for the most up-to-date information.