biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Bugfix] Fix SHM cache initialization (#26427)

Browse Source 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
Cyrus Leung
2025-10-09 17:48:04 +08:00
committed by GitHub
parent dc7976dd9f
commit 4bdf7ac593
30 changed files with 357 additions and 417 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
tests/entrypoints/openai/test_serving_models.py
View File

@@ -25,15 +25,17 @@ LORA_UNLOADING_SUCCESS_MESSAGE = (
async def _async_serving_models_init() -> OpenAIServingModels:
mock_model_config = MagicMock(spec=ModelConfig)
mock_engine_client = MagicMock(spec=EngineClient)
# Set the max_model_len attribute to avoid missing attribute
mock_model_config = MagicMock(spec=ModelConfig)
mock_model_config.max_model_len = 2048
mock_engine_client.model_config = mock_model_config
mock_engine_client.processor = MagicMock()
mock_engine_client.io_processor = MagicMock()
serving_models = OpenAIServingModels(
engine_client=mock_engine_client,
base_model_paths=BASE_MODEL_PATHS,
model_config=mock_model_config,
lora_modules=None,
)
await serving_models.init_static_loras()