fix full graph

Signed-off-by: ganyi <ygan@amd.com>

Author: ganyi1996ppo

vllm/config/scheduler.py

@@ -137,10 +137,6 @@ class SchedulerConfig:
  structured outputs, speculative decoding, and pipeline parallelism.
  """
 
- split_prefill_from_chunk: bool = False
- """Whether to split the prefill request into pure prefill and chunked
- prefill in a single batch."""
-
  def compute_hash(self) -> str:
  """
  WARNING: Whenever a new field is added to this config,

vllm/platforms/rocm.py

@@ -390,11 +390,6 @@ def check_and_update_config(cls, vllm_config: "VllmConfig") -> None:
  ):
  compilation_config.custom_ops.append("+rms_norm")
 
- if envs.VLLM_ROCM_USE_AITER and envs.VLLM_ROCM_USE_AITER_MHA:
- # enable the request reorder if we are using AITER MHA
- # for calculation
- vllm_config.scheduler_config.split_prefill_from_chunk = True
-
  @classmethod
  def verify_model_arch(cls, model_arch: str) -> None:
  if model_arch in _ROCM_UNSUPPORTED_MODELS: