Remove mindie_turbo (#4896)

mindie_turbo is out of data for long time. This PR remove the related register method.

- vLLM version: v0.12.0
- vLLM main:
ad32e3e19c

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>

vllm_ascend/utils.py

@@ -219,19 +219,6 @@ def aligned_16(tensor: torch.Tensor):
     return new_tensor
 
 
-def try_register_lib(lib_name: str, lib_info: str = ""):
-    import importlib
-    import importlib.util
-    try:
-        module_spec = importlib.util.find_spec(lib_name)
-        if module_spec is not None:
-            importlib.import_module(lib_name)
-            if lib_info:
-                logger.info(lib_info)
-    except Exception:
-        pass
-
-
 def enable_custom_op():
     """
     Enable lazy init for vllm_ascend_C to avoid early initialization of CANN's RTS component.

vllm_ascend/worker/worker_v1.py

@@ -53,8 +53,7 @@ from vllm_ascend.distributed.parallel_state import init_ascend_model_parallel
 from vllm_ascend.ops.triton.triton_utils import init_device_properties_triton
 from vllm_ascend.platform import NPUPlatform
 from vllm_ascend.utils import (check_ascend_device_type, enable_sp,
-                               is_enable_nz, register_ascend_customop,
-                               try_register_lib)
+                               is_enable_nz, register_ascend_customop)
 from vllm_ascend.worker.model_runner_v1 import NPUModelRunner
 
 torch._dynamo.trace_rules.clear_lru_cache()  # noqa: E402
@@ -111,11 +110,6 @@ class NPUWorker(WorkerBase):
             except Exception:
                 logger.info("Skip binding cpu.")
 
-        # Try to import mindie_turbo to accelerate vLLM inference.
-        try_register_lib(
-            "mindie_turbo",
-            "MindIE Turbo is installed. vLLM inference will be accelerated with MindIE Turbo."
-        )
         if self.cache_config.cache_dtype == "auto":
             self.cache_dtype = self.model_config.dtype
         else: