@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / b9fd0d7a69
Fork 0

A high-throughput and memory-efficient inference and serving engine for LLMs

Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git

Synced Nov 15, 11:00 UTC

[CI/Build] Fix TPU V1 Test mixed use of & and && across tests (#17968)

Carol Zheng committed May 12, 2025 at 12:06 UTC b9fd0d7a6984bd1b6090f564660c9d1706490700
1 file changed +21 -21
.buildkite/scripts/hardware_ci/run-tpu-v1-test.sh
+21 -21
@@ -26,27 +26,27 @@ docker run --privileged --net host --shm-size=16G -it \
26 && tpu-info \
27 && { \
28 echo TEST_0: Running test_perf.py; \
29 - pytest -s -v /workspace/vllm/tests/tpu/test_perf.py; \
29 + python3 -m pytest -s -v /workspace/vllm/tests/tpu/test_perf.py; \
30 echo TEST_0_EXIT_CODE: \$?; \
31 } & \
32 - && { \
32 + { \
33 echo TEST_1: Running test_compilation.py; \
34 - pytest -s -v /workspace/vllm/tests/tpu/test_compilation.py; \
34 + python3 -m pytest -s -v /workspace/vllm/tests/tpu/test_compilation.py; \
35 echo TEST_1_EXIT_CODE: \$?; \
36 } & \
37 { \
38 echo TEST_2: Running test_basic.py; \
39 - pytest -s -v /workspace/vllm/tests/v1/tpu/test_basic.py; \
39 + python3 -m pytest -s -v /workspace/vllm/tests/v1/tpu/test_basic.py; \
40 echo TEST_2_EXIT_CODE: \$?; \
41 } & \
42 { \
43 echo TEST_3: Running test_accuracy.py::test_lm_eval_accuracy_v1_engine; \
44 - pytest -s -v /workspace/vllm/tests/entrypoints/llm/test_accuracy.py::test_lm_eval_accuracy_v1_engine; \
44 + python3 -m pytest -s -v /workspace/vllm/tests/entrypoints/llm/test_accuracy.py::test_lm_eval_accuracy_v1_engine; \
45 echo TEST_3_EXIT_CODE: \$?; \
46 } & \
47 { \
48 echo TEST_4: Running test_quantization_accuracy.py; \
49 - pytest -s -v /workspace/vllm/tests/tpu/test_quantization_accuracy.py; \
49 + python3 -m pytest -s -v /workspace/vllm/tests/tpu/test_quantization_accuracy.py; \
50 echo TEST_4_EXIT_CODE: \$?; \
51 } & \
52 { \
@@ -56,43 +56,43 @@ docker run --privileged --net host --shm-size=16G -it \
56 } & \
57 { \
58 echo TEST_6: Running test_tpu_model_runner.py; \
59 - pytest -s -v /workspace/vllm/tests/tpu/worker/test_tpu_model_runner.py; \
59 + python3 -m pytest -s -v /workspace/vllm/tests/tpu/worker/test_tpu_model_runner.py; \
60 echo TEST_6_EXIT_CODE: \$?; \
61 } & \
62 - && { \
62 + { \
63 echo TEST_7: Running test_sampler.py; \
64 - pytest -s -v /workspace/vllm/tests/v1/tpu/test_sampler.py; \
64 + python3 -m pytest -s -v /workspace/vllm/tests/v1/tpu/test_sampler.py; \
65 echo TEST_7_EXIT_CODE: \$?; \
66 } & \
67 - && { \
67 + { \
68 echo TEST_8: Running test_topk_topp_sampler.py; \
69 - pytest -s -v /workspace/vllm/tests/v1/tpu/test_topk_topp_sampler.py; \
69 + python3 -m pytest -s -v /workspace/vllm/tests/v1/tpu/test_topk_topp_sampler.py; \
70 echo TEST_8_EXIT_CODE: \$?; \
71 } & \
72 - && { \
72 + { \
73 echo TEST_9: Running test_multimodal.py; \
74 - pytest -s -v /workspace/vllm/tests/v1/tpu/test_multimodal.py; \
74 + python3 -m pytest -s -v /workspace/vllm/tests/v1/tpu/test_multimodal.py; \
75 echo TEST_9_EXIT_CODE: \$?; \
76 } & \
77 - && { \
77 + { \
78 echo TEST_10: Running test_pallas.py; \
79 - pytest -s -v /workspace/vllm/tests/v1/tpu/test_pallas.py; \
79 + python3 -m pytest -s -v /workspace/vllm/tests/v1/tpu/test_pallas.py; \
80 echo TEST_10_EXIT_CODE: \$?; \
81 } & \
82 - && { \
82 + { \
83 echo TEST_11: Running test_struct_output_generate.py; \
84 - pytest -s -v /workspace/vllm/tests/v1/entrypoints/llm/test_struct_output_generate.py; \
84 + python3 -m pytest -s -v /workspace/vllm/tests/v1/entrypoints/llm/test_struct_output_generate.py; \
85 echo TEST_11_EXIT_CODE: \$?; \
86 } & \
87 - && { \
87 + { \
88 echo TEST_12: Running test_moe_pallas.py; \
89 - pytest -s -v /workspace/vllm/tests/tpu/test_moe_pallas.py; \
89 + python3 -m pytest -s -v /workspace/vllm/tests/tpu/test_moe_pallas.py; \
90 echo TEST_12_EXIT_CODE: \$?; \
91 } & \
92 # Disable the TPU LoRA tests until the feature is activated
93 - # && { \
93 + # & { \
94 # echo TEST_13: Running test_moe_pallas.py; \
95 - # pytest -s -v /workspace/vllm/tests/tpu/lora/; \
95 + # python3 -m pytest -s -v /workspace/vllm/tests/tpu/lora/; \
96 # echo TEST_13_EXIT_CODE: \$?; \
97 # } & \
98 wait \