EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

Refactor e2e CI (#2276)

Browse Source 
Refactor E2E CI to make it clear and faster
1. remove some uesless e2e test
2. remove some uesless function
3. Make sure all test runs with VLLMRunner to avoid oom error
4. Make sure all ops test end with torch.empty_cache to avoid oom error
5. run the test one by one to avoid resource limit error


- vLLM version: v0.10.1.1
- vLLM main:
a344a5aa0a

Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
This commit is contained in:
wangxiyuan
2025-09-02 09:02:22 +08:00
committed by GitHub
parent 0df059f41a
commit fef18b60bc
41 changed files with 374 additions and 1757 deletions
Download Patch File Download Diff File Expand all files Collapse all files

50
.github/workflows/vllm_ascend_test.yaml vendored
View File

@@ -191,27 +191,29 @@ jobs:
VLLM_WORKER_MULTIPROC_METHOD: spawn
VLLM_USE_MODELSCOPE: True
run: |
pytest -sv tests/e2e/singlecard/test_offline_inference.py
# pytest -sv tests/e2e/singlecard/test_ilama_lora.py
pytest -sv tests/e2e/singlecard/test_guided_decoding.py
# We found that if running aclgraph tests in batch, it will cause AclmdlRICaptureBegin error. So we run
# the test separately.
pytest -sv tests/e2e/singlecard/test_aclgraph.py
pytest -sv tests/e2e/singlecard/test_ascend_scheduler.py
pytest -sv tests/e2e/singlecard/test_camem.py
pytest -sv tests/e2e/singlecard/test_chunked.py
pytest -sv tests/e2e/singlecard/test_embedding.py
pytest -sv tests/e2e/singlecard/test_guided_decoding.py
# TODO: Fix lora accuracy error
# pytest -sv tests/e2e/singlecard/test_ilama_lora.py
pytest -sv tests/e2e/singlecard/test_profile_execute_duration.py
pytest -sv tests/e2e/singlecard/test_quantization.py
pytest -sv tests/e2e/singlecard/test_sampler.py
pytest -sv tests/e2e/singlecard/test_vlm.py
# ------------------------------------ v1 spec decode test ------------------------------------ #
pytest -sv tests/e2e/singlecard/spec_decode_v1/test_v1_mtp_correctness.py
# TODO: revert me when test_v1_spec_decode.py::test_ngram_correctness is fixed
pytest -sv tests/e2e/singlecard/spec_decode_v1/test_v1_spec_decode.py
# All other tests, ignore: 310p test, accuracy test.
pytest -sv tests/e2e/singlecard/ \
--ignore=tests/e2e/singlecard/test_offline_inference.py \
--ignore=tests/e2e/singlecard/test_ilama_lora.py \
--ignore=tests/e2e/singlecard/test_guided_decoding.py \
--ignore=tests/e2e/singlecard/test_camem.py \
--ignore=tests/e2e/singlecard/test_embedding.py \
--ignore=tests/e2e/singlecard/spec_decode_v1/test_v1_mtp_correctness.py \
--ignore=tests/e2e/singlecard/spec_decode_v1/test_v1_spec_decode.py \
--ignore=tests/e2e/singlecard/test_offline_inference_310p.py
pytest -sv tests/e2e/singlecard/ops/
e2e-2-cards:
needs: [e2e]
if: ${{ needs.e2e.result == 'success' }}
@@ -273,17 +275,23 @@ jobs:
VLLM_WORKER_MULTIPROC_METHOD: spawn
VLLM_USE_MODELSCOPE: True
run: |
pytest -sv tests/e2e/multicard/test_data_parallel.py
pytest -sv tests/e2e/multicard/test_expert_parallel.py
# external_launcher test is not stable enough. Fix it later
# pytest -sv tests/e2e/multicard/test_external_launcher.py
pytest -sv tests/e2e/multicard/test_fused_moe_allgather_ep.py
# pytest -sv tests/e2e/multicard/test_ilama_lora_tp2.py
# Fixme: run VLLM_USE_MODELSCOPE=True pytest -sv tests/e2e/multicard/test_offline_inference_distributed.py will raise error.
# To avoid oom, we need to run the test in a single process.
pytest -sv tests/e2e/multicard/test_offline_inference_distributed.py::test_models_distributed_DeepSeek_multistream_moe
pytest -sv tests/e2e/multicard/test_offline_inference_distributed.py::test_models_distributed_QwQ
pytest -sv tests/e2e/multicard/test_offline_inference_distributed.py::test_models_distributed_DeepSeekV3_dbo
pytest -sv tests/e2e/multicard/test_offline_inference_distributed.py::test_models_distributed_DeepSeek_multistream_moe
#pytest -sv tests/e2e/multicard/test_offline_inference_distributed.py::test_models_distributed_pangu
#pytest -sv tests/e2e/multicard/test_offline_inference_distributed.py::test_models_distributed_Qwen3_W8A8
pytest -sv tests/e2e/multicard/test_offline_inference_distributed.py::test_models_distributed_Qwen3_W4A8DYNAMIC
pytest -sv tests/e2e/multicard/test_offline_inference_distributed.py::test_models_distributed_DeepSeek_W4A8DYNAMIC
pytest -sv tests/e2e/multicard/test_offline_inference_distributed.py::test_sp_for_qwen3_moe
pytest -sv tests/e2e/multicard/test_data_parallel.py
pytest -sv tests/e2e/multicard/ --ignore=tests/e2e/multicard/test_ilama_lora_tp2.py \
--ignore=tests/e2e/multicard/test_offline_inference_distributed.py \
--ignore=tests/e2e/multicard/test_data_parallel.py \
--ignore=tests/e2e/multicard/test_offline_inference_310p.py
#pytest -sv tests/e2e/multicard/test_pipeline_parallel.py
#pytest -sv tests/e2e/multicard/test_prefix_caching.py
#pytest -sv tests/e2e/multicard/test_qwen3_moe.py
#pytest -sv tests/e2e/multicard/test_torchair_graph_mode.py

6
.github/workflows/vllm_ascend_test_310p.yaml vendored
View File

@@ -111,7 +111,7 @@ jobs:
PYTORCH_NPU_ALLOC_CONF: max_split_size_mb:256
run: |
if [[ "${{ matrix.os }}" == "linux-aarch64-310p-1" ]]; then
pytest -sv tests/e2e/singlecard/test_offline_inference_310p.py
pytest -sv tests/e2e/310p/test_offline_inference_310p.py
else
pytest -sv tests/e2e/multicard/test_offline_inference_310p.py
fi
pytest -sv tests/e2e/310p/test_offline_inference_parallel_310p.py
fi