You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: .github/workflows/docker-tag-monitor.yml
+12-9Lines changed: 12 additions & 9 deletions
Original file line number
Diff line number
Diff line change
@@ -159,9 +159,11 @@ jobs:
159
159
echo ""
160
160
echo "1. Update image tags in \`.github/configs/nvidia-master.yaml\` and/or \`.github/configs/amd-master.yaml\`"
161
161
echo "2. Add entries to \`perf-changelog.yaml\` documenting the version changes"
162
-
echo "3. Create a PR with these changes"
162
+
echo "3. Create separate PRs grouped by framework and image family, and link each PR."
163
163
echo ""
164
-
echo "Focus on updating single-node configurations. For each framework, check if there are multiple CUDA/ROCm versions available and choose appropriately based on current usage patterns in the configs."
164
+
echo "Group by framework plus CUDA/ROCm image family (for example sglang-cuda, sglang-rocm, vllm-cuda, vllm-rocm, atom, trt), not by individual GPU. Split by GPU only when an update is genuinely hardware-specific."
165
+
echo ""
166
+
echo "Focus on updating single-node configurations. For each framework/image family, check if there are multiple CUDA/ROCm versions available and choose appropriately based on current usage patterns in the configs."
0 commit comments