xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-16 05:55:01 +08:00
Code Issues Packages Projects Releases Wiki Activity

[TPU] Skip hanging tests (#19115)

Browse Source 
Signed-off-by: Siyuan Liu <lsiyuan@google.com>
This commit is contained in:
Siyuan Liu 2025-06-04 01:43:00 -07:00 committed by GitHub
parent 3336c8cfbe
commit 8e972d9c44
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 9 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
.buildkite/scripts/hardware_ci/run-tpu-v1-test.sh
View File

@ -150,7 +150,7 @@ run_and_track_test 9 "test_multimodal.py" \
run_and_track_test 10 "test_pallas.py" \ run_and_track_test 10 "test_pallas.py" \
"python3 -m pytest -s -v /workspace/vllm/tests/v1/tpu/test_pallas.py" "python3 -m pytest -s -v /workspace/vllm/tests/v1/tpu/test_pallas.py"
run_and_track_test 11 "test_struct_output_generate.py" \ run_and_track_test 11 "test_struct_output_generate.py" \
"python3 -m pytest -s -v /workspace/vllm/tests/v1/entrypoints/llm/test_struct_output_generate.py" "python3 -m pytest -s -v /workspace/vllm/tests/v1/entrypoints/llm/test_struct_output_generate.py -k 'not test_structured_output_with_reasoning_matrices'"
run_and_track_test 12 "test_moe_pallas.py" \ run_and_track_test 12 "test_moe_pallas.py" \
"python3 -m pytest -s -v /workspace/vllm/tests/tpu/test_moe_pallas.py" "python3 -m pytest -s -v /workspace/vllm/tests/tpu/test_moe_pallas.py"
run_and_track_test 13 "test_lora.py" \ run_and_track_test 13 "test_lora.py" \

13
tests/v1/tpu/test_spmd_model_weight_loading.py
View File

@ -45,11 +45,14 @@ def _get_spmd_mesh():
return MESH return MESH
@pytest.mark.parametrize("model", [ @pytest.mark.parametrize(
"Qwen/Qwen2-1.5B-Instruct", "model",
"meta-llama/Llama-3.1-8B-Instruct", [
"meta-llama/Llama-3.1-70B-Instruct", "Qwen/Qwen2-1.5B-Instruct",
]) # Skip large models due to CI runner disk space limitations
# "meta-llama/Llama-3.1-8B-Instruct",
# "meta-llama/Llama-3.1-70B-Instruct",
])
def test_tpu_model_loader(model): def test_tpu_model_loader(model):
# Skip the 70B test if there are less than 8 chips # Skip the 70B test if there are less than 8 chips
# TODO: Query using torch xla API, the query API is not working # TODO: Query using torch xla API, the query API is not working