xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-10 02:05:01 +08:00
Code Issues Packages Projects Releases Wiki Activity

[easy] Print number of needed GPUs in skip message (#17594)

Browse Source 
Signed-off-by: rzou <zou3519@gmail.com>
This commit is contained in:
Richard Zou 2025-05-02 18:27:43 -04:00 committed by GitHub
parent 9352cdb56d
commit b90b0852e9
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 2 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
tests/compile/test_basic_correctness.py
View File

@ -103,7 +103,8 @@ def test_compile_correctness(
method = test_setting.method method = test_setting.method
fullgraph = test_setting.fullgraph fullgraph = test_setting.fullgraph
if cuda_device_count_stateless() != pp_size * tp_size: if cuda_device_count_stateless() != pp_size * tp_size:
pytest.skip("Not correct CUDA devices for the test.") pytest.skip(f"Need exactly {pp_size}*{tp_size} CUDA gpus but got "
f"{cuda_device_count_stateless()}")
with monkeypatch.context() as m: with monkeypatch.context() as m:
m.setenv("VLLM_ATTENTION_BACKEND", attn_backend) m.setenv("VLLM_ATTENTION_BACKEND", attn_backend)