fix: add intel-xe accelerator environment variable support

- add VLLM_WORKER_MULTIPROC_METHOD: spawn for Intel-xe
- bump version
Signed-off-by: Wen Zhou <wenzhou@redhat.com>

Author: zdtsw

charts/llm-d-modelservice/Chart.yaml

@@ -13,7 +13,7 @@ type: application
 # This is the chart version. This version number should be incremented each time you make changes
 # to the chart and its templates, including the app version.
 # Versions are expected to follow Semantic Versioning (https://semver.org/)
-version: "v0.4.2"
+version: "v0.4.3"
 # This is the version number of the application being deployed. This version number should be
 # incremented each time you make changes to the application. Versions are not expected to
 # follow Semantic Versioning. They should reflect the version the application is using.

charts/llm-d-modelservice/values.schema.json

@@ -14,7 +14,7 @@
                     "type": "boolean"
                 },
                 "env": {
-                    "additionalProperties": false,
+                    "additionalProperties": true,
                     "description": "Environment variables specific to accelerator types",
                     "properties": {
                         "intel-i915": {

charts/llm-d-modelservice/values.schema.tmpl.json

@@ -14,7 +14,7 @@
           "type": "boolean"
         },
         "env": {
-          "additionalProperties": false,
+          "additionalProperties": true,
           "description": "Environment variables specific to accelerator types",
           "properties": {
             "intel-i915": {

charts/llm-d-modelservice/values.yaml

@@ -74,7 +74,7 @@ modelArtifacts:
 multinode: false
 
 # Global accelerator configuration
-# Supported types: nvidia, intel-i915, intel-xe, intel-gaudi, amd, google
+# Supported types: nvidia, intel-i915, intel-xe, intel-gaudi, amd, google, cpu
 accelerator:
   # Type of accelerator to use
   type: nvidia
@@ -101,6 +101,9 @@ accelerator:
         value: "1"
       - name: VLLM_WORKER_MULTIPROC_METHOD
         value: "spawn"
+    intel-xe:
+      - name: VLLM_WORKER_MULTIPROC_METHOD
+        value: "spawn"
   # ResourceClaimTemplate configurations for DRA (used when dra: true)
   # Each accelerator type can have its own claim template configuration
   resourceClaimTemplates:

examples/output-cpu.yaml

@@ -6,7 +6,7 @@ kind: ServiceAccount
 metadata:
   name: cpu-sim-llm-d-modelservice
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.2
+    helm.sh/chart: llm-d-modelservice-v0.4.3
     app.kubernetes.io/version: "v0.3.0"
     app.kubernetes.io/managed-by: Helm
 ---
@@ -16,7 +16,7 @@ kind: Deployment
 metadata:
   name: cpu-sim-llm-d-modelservice-decode
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.2
+    helm.sh/chart: llm-d-modelservice-v0.4.3
     app.kubernetes.io/version: "v0.3.0"
     app.kubernetes.io/managed-by: Helm
 spec:
@@ -106,7 +106,7 @@ kind: Deployment
 metadata:
   name: cpu-sim-llm-d-modelservice-prefill
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.2
+    helm.sh/chart: llm-d-modelservice-v0.4.3
     app.kubernetes.io/version: "v0.3.0"
     app.kubernetes.io/managed-by: Helm
 spec:

examples/output-dra.yaml

@@ -6,7 +6,7 @@ kind: ServiceAccount
 metadata:
   name: dra-llm-d-modelservice
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.2
+    helm.sh/chart: llm-d-modelservice-v0.4.3
     app.kubernetes.io/version: "v0.3.0"
     app.kubernetes.io/managed-by: Helm
 ---
@@ -16,7 +16,7 @@ kind: Deployment
 metadata:
   name: dra-llm-d-modelservice-decode
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.2
+    helm.sh/chart: llm-d-modelservice-v0.4.3
     app.kubernetes.io/version: "v0.3.0"
     app.kubernetes.io/managed-by: Helm
 spec:
@@ -116,7 +116,7 @@ kind: ResourceClaimTemplate
 metadata:
   name: intel-gaudi-claim-template-decode
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.2
+    helm.sh/chart: llm-d-modelservice-v0.4.3
     app.kubernetes.io/version: "v0.3.0"
     app.kubernetes.io/managed-by: Helm
     llm-d.ai/role: decode

examples/output-gaudi.yaml

@@ -6,7 +6,7 @@ kind: ServiceAccount
 metadata:
   name: gaudi-llm-d-modelservice
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.2
+    helm.sh/chart: llm-d-modelservice-v0.4.3
     app.kubernetes.io/version: "v0.3.0"
     app.kubernetes.io/managed-by: Helm
 ---
@@ -16,7 +16,7 @@ kind: Deployment
 metadata:
   name: gaudi-llm-d-modelservice-decode
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.2
+    helm.sh/chart: llm-d-modelservice-v0.4.3
     app.kubernetes.io/version: "v0.3.0"
     app.kubernetes.io/managed-by: Helm
 spec:

examples/output-heterogeneous-pd.yaml

@@ -6,7 +6,7 @@ kind: ServiceAccount
 metadata:
   name: heterogeneous-pd-llm-d-modelservice
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.2
+    helm.sh/chart: llm-d-modelservice-v0.4.3
     app.kubernetes.io/version: "v0.3.0"
     app.kubernetes.io/managed-by: Helm
 ---
@@ -16,7 +16,7 @@ kind: Deployment
 metadata:
   name: heterogeneous-pd-llm-d-modelservice-decode
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.2
+    helm.sh/chart: llm-d-modelservice-v0.4.3
     app.kubernetes.io/version: "v0.3.0"
     app.kubernetes.io/managed-by: Helm
 spec:
@@ -132,7 +132,7 @@ kind: Deployment
 metadata:
   name: heterogeneous-pd-llm-d-modelservice-prefill
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.2
+    helm.sh/chart: llm-d-modelservice-v0.4.3
     app.kubernetes.io/version: "v0.3.0"
     app.kubernetes.io/managed-by: Helm
 spec:
@@ -226,7 +226,7 @@ kind: ResourceClaimTemplate
 metadata:
   name: nvidia-claim-template-decode
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.2
+    helm.sh/chart: llm-d-modelservice-v0.4.3
     app.kubernetes.io/version: "v0.3.0"
     app.kubernetes.io/managed-by: Helm
     llm-d.ai/role: decode

examples/output-pd-mnnvl.yaml

@@ -6,7 +6,7 @@ kind: ServiceAccount
 metadata:
   name: pd-mnnvl-llm-d-modelservice
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.2
+    helm.sh/chart: llm-d-modelservice-v0.4.3
     app.kubernetes.io/version: "v0.3.0"
     app.kubernetes.io/managed-by: Helm
 ---
@@ -16,7 +16,7 @@ kind: Deployment
 metadata:
   name: pd-mnnvl-llm-d-modelservice-decode
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.2
+    helm.sh/chart: llm-d-modelservice-v0.4.3
     app.kubernetes.io/version: "v0.3.0"
     app.kubernetes.io/managed-by: Helm
 spec:
@@ -132,7 +132,7 @@ kind: Deployment
 metadata:
   name: pd-mnnvl-llm-d-modelservice-prefill
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.2
+    helm.sh/chart: llm-d-modelservice-v0.4.3
     app.kubernetes.io/version: "v0.3.0"
     app.kubernetes.io/managed-by: Helm
 spec:

examples/output-pd.yaml

@@ -6,7 +6,7 @@ kind: ServiceAccount
 metadata:
   name: pd-llm-d-modelservice
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.2
+    helm.sh/chart: llm-d-modelservice-v0.4.3
     app.kubernetes.io/version: "v0.3.0"
     app.kubernetes.io/managed-by: Helm
 ---
@@ -16,7 +16,7 @@ kind: Deployment
 metadata:
   name: pd-llm-d-modelservice-decode
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.2
+    helm.sh/chart: llm-d-modelservice-v0.4.3
     app.kubernetes.io/version: "v0.3.0"
     app.kubernetes.io/managed-by: Helm
 spec:
@@ -128,7 +128,7 @@ kind: Deployment
 metadata:
   name: pd-llm-d-modelservice-prefill
   labels:
-    helm.sh/chart: llm-d-modelservice-v0.4.2
+    helm.sh/chart: llm-d-modelservice-v0.4.3
     app.kubernetes.io/version: "v0.3.0"
     app.kubernetes.io/managed-by: Helm
 spec: