From my point of view, this can be backported if needed, is just increasing the capacity of existing features, as the limits were hit in specific deployments, no new feature added. Like we had to do in the past to increase internal buffer sizes to cope with larger values.