ebable sleep mode test

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

Author: Potabk

.github/workflows/vllm_ascend_test.yaml

@@ -127,7 +127,9 @@ jobs:
             pytest -sv tests/singlecard/test_scheduler.py
             # guided decoding doesn't work, fix it later
             # pytest -sv tests/singlecard/test_guided_decoding.py.py
-            pytest -sv tests/singlecard/ --ignore=tests/singlecard/test_offline_inference.py --ignore=tests/singlecard/test_scheduler.py --ignore=tests/singlecard/test_guided_decoding.py
+            pytest -sv tests/singlecard/test_camem.py
+            pytest -sv tests/singlecard/test_ilama_lora.py
+            pytest -sv tests/singlecard/test_pyhccl.py
           else
             pytest -sv tests/multicard/test_ilama_lora_tp2.py
             # Fixme: run VLLM_USE_MODELSCOPE=True pytest -sv tests/multicard/test_offline_inference_distributed.py will raise error.

tests/singlecard/test_camem.py

@@ -16,6 +16,8 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 #
+import os
+
 import pytest
 import torch
 from vllm import LLM, SamplingParams
@@ -24,7 +26,11 @@
 from tests.utils import fork_new_process_for_each_test
 from vllm_ascend.device_allocator.camem import CaMemAllocator
 
+if os.getenv("VLLM_USE_V1") == "1":
+    pytest.skip("Skip in vllm v1", allow_module_level=True)
+
 
+@fork_new_process_for_each_test
 def test_basic_camem():
     # some tensors from default memory pool
     shape = (1024, 1024)
@@ -57,7 +63,6 @@ def test_basic_camem():
     assert torch.allclose(output, torch.ones_like(output) * 3)
 
 
-@pytest.mark.skipif(True, reason="test failed, should be fixed later")
 @fork_new_process_for_each_test
 def test_end_to_end():
     free, total = torch.npu.mem_get_info()