Drop ascend scheduler (#4623)

It's safe to drop ascend scheduler now. The related test and doc has
been removed already


- vLLM version: v0.12.0
- vLLM main:
vllm-project/vllm@ad32e3e

Signed-off-by: wangxiyuan <[email protected]>

Author: wangxiyuan

.github/workflows/vllm_ascend_test_pr_light.yaml

@@ -139,7 +139,6 @@ jobs:
             --ignore tests/ut/kv_connector/test_remote_prefill_lifecycle.py \
             --ignore tests/ut/kv_connector/test_remote_decode_lifecycle.py \
             --ignore tests/ut/kv_connector/test_llmdatadist_connector.py \
-            --ignore tests/ut/ops/test_linear.py \
             --ignore tests/ut/core/test_scheduler_dynamic_batch.py
 
       - name: Upload coverage to Codecov

tests/ut/ops/test_linear.py

@@ -99,7 +99,7 @@ def test_oproj_tp(self):
 
         ascend_config._ASCEND_CONFIG = MagicMock()
         ascend_config._ASCEND_CONFIG.oproj_tensor_parallel_size = 2
-        ascend_config._ASCEND_CONFIG.ascend_scheduler_config.enabled = False
+        ascend_config._ASCEND_CONFIG.recompute_scheduler_enable = False
 
         linear = AscendRowParallelLinear(
             input_size=16,

tests/ut/ops/test_vocab_parallel_embedding.py

@@ -209,12 +209,7 @@ def setUp(self):
                 return_value=torch.randn(1, self.vocab_size)),
             patch(
                 "vllm_ascend.ops.vocab_parallel_embedding.get_lmhead_tp_group.all_gather",
-                return_value=torch.randn(1, self.vocab_size)),
-            patch(
-                "vllm_ascend.core.schedule_config.AscendSchedulerConfig.initialize_from_config",
-                return_value=MagicMock(max_num_batched_tokens=1000,
-                                       max_model_len=512,
-                                       enable_chunked_prefill=False))
+                return_value=torch.randn(1, self.vocab_size))
         ]
 
         for p in self.patches:

tests/ut/test_platform.py

@@ -32,7 +32,6 @@ def mock_vllm_config():
     def mock_vllm_ascend_config():
         mock_ascend_config = MagicMock()
         mock_ascend_config.torchair_graph_config.enabled = False
-        mock_ascend_config.ascend_scheduler_config.enabled = False
         mock_ascend_config.enable_shared_expert_dp = False
         return mock_ascend_config
 

vllm_ascend/ascend_config.py

@@ -45,11 +45,6 @@ def __init__(self, vllm_config):
         self.ascend_compilation_config = AscendCompilationConfig(
             **ascend_compilation_config)
 
-        ascend_scheduler_config = additional_config.get(
-            "ascend_scheduler_config", {})
-        self.ascend_scheduler_config = AscendSchedulerConfig(
-            ascend_scheduler_config)
-
         # Dump / PrecisionDebugger configuration
         dump_config_path = additional_config.get("dump_config", None)
         self.dump_config = DumpConfig(dump_config_path)
@@ -255,20 +250,6 @@ def __init__(self, torchair_graph_config, vllm_config, additional_config):
             )
 
 
-class AscendSchedulerConfig:
-    """
-    Configuration Object for ascend_scheduler_config from additional_config
-    """
-
-    def __init__(self, ascend_scheduler_config: dict):
-        self.enabled = ascend_scheduler_config.get("enabled", False)
-        # Ascend scheduler is based on vllm v0 scheduler, so we should support
-        # all vllm v0 scheduler configs as well.
-        for k, v in ascend_scheduler_config.items():
-            if not hasattr(self, k):
-                setattr(self, k, v)
-
-
 class DumpConfig:
     """
     Configuration object for dump/PrecisionDebugger settings.