[Misc] Further clean up chunked prefill and prefix caching init

vllm/config/cache.py

@@ -73,8 +73,8 @@ class CacheConfig:
     sliding_window: int | None = None
     """Sliding window size for the KV cache. This is primarily set in
     `ModelConfig` and that value should be manually duplicated here."""
-    enable_prefix_caching: bool | None = None
-    """Whether to enable prefix caching. Enabled by default for V1."""
+    enable_prefix_caching: bool = True
+    """Whether to enable prefix caching."""
     prefix_caching_hash_algo: PrefixCachingHashAlgo = "sha256"
     """Set the hash algorithm for prefix caching:\n
     - "sha256" uses Pickle for object serialization before hashing.\n

vllm/engine/arg_utils.py

@@ -1979,6 +1979,16 @@ def _set_default_args(
                 "%s chunked prefill by default",
                 "Enabling" if default_chunked_prefill else "Disabling",
             )
+        elif (
+            model_config.runner_type == "generate"
+            and not self.enable_chunked_prefill
+            and default_chunked_prefill
+        ):
+            logger.warning(
+                "This model does not officially support disabling chunked prefill. "
+                "Disabling this manually may cause the engine to crash "
+                "or produce incorrect outputs.",
+            )
         elif (
             model_config.runner_type == "pooling"
             and self.enable_chunked_prefill

vllm/v1/core/sched/scheduler.py

@@ -179,7 +179,7 @@ def __init__(
         self.kv_cache_manager = KVCacheManager(
             kv_cache_config=kv_cache_config,
             max_model_len=self.max_model_len,
-            enable_caching=bool(self.cache_config.enable_prefix_caching),
+            enable_caching=self.cache_config.enable_prefix_caching,
             use_eagle=self.use_eagle,
             log_stats=self.log_stats,
             enable_kv_cache_events=self.enable_kv_cache_events,