| 1 | group: Hardware |
| 2 | depends_on: ~ |
| 3 | steps: |
| 4 | - label: "Ascend NPU Test" |
| 5 | key: ascend-npu-test |
| 6 | soft_fail: true |
| 7 | no_plugin: true |
| 8 | device: ascend_npu |
| 9 | timeout_in_minutes: 30 |
| 10 | source_file_dependencies: |
| 11 | - vllm/ |
| 12 | - .buildkite/scripts/hardware_ci/run-npu-test.sh |
| 13 | - "!vllm/platforms/cuda.py" |
| 14 | - "!vllm/platforms/xpu.py" |
| 15 | - "!vllm/platforms/tpu.py" |
| 16 | - "!vllm/platforms/cpu.py" |
| 17 | - "!vllm/platforms/zen_cpu.py" |
| 18 | - "!vllm/cute_utils/" |
| 19 | - "!vllm/tilelang_utils/" |
| 20 | - "!vllm/device_allocator/" |
| 21 | - "!vllm/vllm_flash_attn/" |
| 22 | - "!vllm/kernels/" |
| 23 | - "!vllm/model_executor/kernels/" |
| 24 | - "!vllm/engine/" |
| 25 | - "!vllm/entrypoints/" |
| 26 | - "!vllm/ray/" |
| 27 | - "!vllm/v1/attention/backends/flash_attn.py" |
| 28 | - "!vllm/v1/attention/backends/flash_attn_diffkv.py" |
| 29 | - "!vllm/v1/attention/backends/flashinfer.py" |
| 30 | - "!vllm/v1/attention/backends/triton_attn.py" |
| 31 | - "!vllm/v1/attention/backends/triton_attn_diffkv.py" |
| 32 | - "!vllm/v1/attention/backends/rocm_attn.py" |
| 33 | - "!vllm/v1/attention/backends/rocm_aiter_fa.py" |
| 34 | - "!vllm/v1/attention/backends/rocm_aiter_unified_attn.py" |
| 35 | - "!vllm/v1/attention/backends/flex_attention.py" |
| 36 | - "!vllm/v1/attention/backends/hpc_attn.py" |
| 37 | - "!vllm/v1/attention/backends/mamba1_attn.py" |
| 38 | - "!vllm/v1/attention/backends/mamba2_attn.py" |
| 39 | - "!vllm/v1/attention/backends/mamba_attn.py" |
| 40 | - "!vllm/v1/attention/backends/short_conv_attn.py" |
| 41 | - "!vllm/v1/attention/backends/turboquant_attn.py" |
| 42 | - "!vllm/v1/attention/backends/cpu_attn.py" |
| 43 | - "!vllm/v1/attention/backends/fa_utils.py" |
| 44 | - "!vllm/distributed/device_communicators/pynccl.py" |
| 45 | - "!vllm/distributed/device_communicators/pynccl_wrapper.py" |
| 46 | - "!vllm/distributed/device_communicators/pynccl_allocator.py" |
| 47 | - "!vllm/distributed/device_communicators/xpu_communicator.py" |
| 48 | - "!vllm/distributed/device_communicators/cpu_communicator.py" |
| 49 | - "!vllm/distributed/device_communicators/cuda_wrapper.py" |
| 50 | - "!vllm/distributed/device_communicators/flashinfer_all_reduce.py" |
| 51 | - "!vllm/distributed/device_communicators/aiter_custom_all_reduce.py" |
| 52 | - "!vllm/distributed/device_communicators/custom_all_reduce.py" |
| 53 | - "!vllm/distributed/device_communicators/quick_all_reduce.py" |
| 54 | - "!vllm/distributed/device_communicators/all2all.py" |
| 55 | - "!vllm/distributed/device_communicators/all_reduce_utils.py" |
| 56 | - "!vllm/distributed/device_communicators/shm_object_storage.py" |
| 57 | - "!vllm/distributed/device_communicators/symm_mem.py" |
| 58 | - "!vllm/distributed/device_communicators/mnnvl_compat.py" |
| 59 | - "!vllm/distributed/device_communicators/ray_communicator.py" |
| 60 | - "!vllm/third_party/flashmla/" |
| 61 | - "!vllm/third_party/pynvml.py" |
| 62 | - "!vllm/_custom_ops.py" |
| 63 | - "!vllm/_xpu_ops.py" |
| 64 | - "!vllm/_aiter_ops" |
| 65 | - "!vllm/collect_env.py" |
| 66 | - "!vllm/connections.py" |
| 67 | - "!vllm/env_override.py" |
| 68 | - "!vllm/logits_process.py" |
| 69 | - "!vllm/logprobs.py" |
| 70 | - "!vllm/model_inspection.py" |
| 71 | - "!vllm/outputs.py" |
| 72 | - "!vllm/scalar_type.py" |
| 73 | - "!vllm/scripts.py" |
| 74 | - "!vllm/version.py" |
| 75 | - "!vllm/assets/" |
| 76 | - "!vllm/benchmarks/" |
| 77 | - "!vllm/inputs/" |
| 78 | - "!vllm/ir/" |
| 79 | - "!vllm/parser/" |
| 80 | - "!vllm/plugins/" |
| 81 | - "!vllm/reasoning/" |
| 82 | - "!vllm/renderers/" |
| 83 | - "!vllm/tokenizers/" |
| 84 | - "!vllm/tool_parsers/" |
| 85 | - "!vllm/tracing/" |
| 86 | - "!vllm/usage/" |
| 87 | - "!tests/" |
| 88 | - "!benchmarks/" |
| 89 | - "!docs/" |
| 90 | - "!examples/" |
| 91 | - "!scripts/" |
| 92 | - "!vllm/model_executor/layers/quantization/auto_awq.py" |
| 93 | - "!vllm/model_executor/layers/quantization/auto_gptq.py" |
| 94 | - "!vllm/model_executor/layers/quantization/inc/" |
| 95 | - "!vllm/model_executor/layers/quantization/awq_triton.py" |
| 96 | - "!vllm/model_executor/layers/quantization/experts_int8.py" |
| 97 | - "!vllm/model_executor/layers/quantization/fbgemm_fp8.py" |
| 98 | - "!vllm/model_executor/layers/quantization/fp_quant.py" |
| 99 | - "!vllm/model_executor/layers/quantization/humming.py" |
| 100 | - "!vllm/model_executor/layers/quantization/moe_wna16.py" |
| 101 | - "!vllm/model_executor/layers/quantization/qutlass_utils.py" |
| 102 | - "!vllm/model_executor/layers/quantization/quark/" |
| 103 | - "!vllm/model_executor/layers/quantization/turboquant/" |
| 104 | commands: |
| 105 | - bash .buildkite/scripts/hardware_ci/run-npu-test.sh |