Disable chunked prefill and/or prefix caching when MLA is enabled (vl…

…lm-project#12642)

From @mgoin in vllm-project#12638

I cannot push to that branch, therefore a new PR to unblock release.

---------

Signed-off-by: mgoin <michael@neuralmagic.com>
Signed-off-by: simon-mo <simon.mo@hey.com>
Co-authored-by: mgoin <michael@neuralmagic.com>

vllm/config.py

@@ -3252,6 +3252,16 @@ def __post_init__(self):
 
         current_platform.check_and_update_config(self)
 
+        # If MLA is enabled, force disable chunked prefill and prefix caching
+        if self.model_config and self.model_config.use_mla:
+            logger.info("MLA is enabled; forcing chunked prefill and prefix "
+                        "caching to be disabled.")
+            self.scheduler_config.enable_chunked_prefill = False
+            self.scheduler_config.chunked_prefill_enabled = False
+
+            if self.cache_config is not None:
+                self.cache_config.enable_prefix_caching = False
+
         if not self.instance_id:
             self.instance_id = random_uuid()[:5]