kanshan/TensorRT-LLMs
1
0
Fork 0
mirror of https://github.com/NVIDIA/TensorRT-LLM.git synced 2026-01-14 06:27:45 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

[TRTLLM-6471] Infra: unwaive nixl tests and some disagg-serve tests (#6095)

Browse Source 
Signed-off-by: Bo Deng <deemod@nvidia.com>
This commit is contained in:
Bo Deng 2025-07-19 00:48:44 +08:00 committed by GitHub
parent 07e8813984
commit 2c6fa145ee
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
5 changed files with 11 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
tests/integration/defs/cpp/test_multi_gpu.py
View File

@ -108,8 +108,6 @@ def run_cache_transceiver_tests(build_dir: _pl.Path,
env=mgpu_env,
timeout=timeout)
# TODO: Re-enable it after the NIXL backend has stabilized.
'''
# Nixl transfer agent tests
new_env = get_multi_gpu_env(kv_cache_type=KVCacheType.NIXL)
@ -125,7 +123,6 @@ def run_cache_transceiver_tests(build_dir: _pl.Path,
cwd=tests_dir,
env=new_env,
timeout=600)
'''
def run_llama_executor_leader_tests(build_dir: _pl.Path, timeout=1500):

1
tests/integration/test_lists/qa/examples_test_list.txt
View File

@ -591,6 +591,7 @@ disaggregated/test_disaggregated.py::test_disaggregated_single_gpu_with_mpirun_t
disaggregated/test_disaggregated.py::test_disaggregated_cuda_graph[TinyLlama-1.1B-Chat-v1.0]
disaggregated/test_disaggregated.py::test_disaggregated_deepseek_v3_lite_fp8_mpi[DeepSeek-V3-Lite-fp8]
disaggregated/test_disaggregated.py::test_disaggregated_deepseek_v3_lite_fp8_ucx[DeepSeek-V3-Lite-fp8]
disaggregated/test_disaggregated.py::test_disaggregated_deepseek_v3_lite_fp8_nixl[DeepSeek-V3-Lite-fp8]
disaggregated/test_disaggregated.py::test_disaggregated_deepseek_v3_lite_fp8_attention_dp[DeepSeek-V3-Lite-fp8]
disaggregated/test_disaggregated.py::test_disaggregated_deepseek_v3_lite_fp8_attention_dp_one[DeepSeek-V3-Lite-fp8]
disaggregated/test_disaggregated.py::test_disaggregated_deepseek_v3_lite_fp8_attention_dp_one_mtp[DeepSeek-V3-Lite-fp8]

1
tests/integration/test_lists/qa/llm_sanity_test.txt
View File

@ -61,6 +61,7 @@ disaggregated/test_disaggregated.py::test_disaggregated_deepseek_v3_lite_fp8_att
disaggregated/test_disaggregated.py::test_disaggregated_deepseek_v3_lite_fp8_attention_dp[DeepSeek-V3-Lite-fp8]
disaggregated/test_disaggregated.py::test_disaggregated_deepseek_v3_lite_fp8_ucx[DeepSeek-V3-Lite-fp8]
disaggregated/test_disaggregated.py::test_disaggregated_deepseek_v3_lite_fp8_mpi[DeepSeek-V3-Lite-fp8]
disaggregated/test_disaggregated.py::test_disaggregated_deepseek_v3_lite_fp8_nixl[DeepSeek-V3-Lite-fp8]
disaggregated/test_disaggregated.py::test_disaggregated_load_balance[TinyLlama-1.1B-Chat-v1.0]
disaggregated/test_disaggregated.py::test_disaggregated_cache_aware_balance[TinyLlama-1.1B-Chat-v1.0]
disaggregated/test_disaggregated.py::test_disaggregated_trtllm_sampler[TinyLlama-1.1B-Chat-v1.0]

2
tests/integration/test_lists/test-db/l0_dgx_b200.yml
View File

@ -64,3 +64,5 @@ l0_dgx_b200:
- accuracy/test_llm_api_pytorch.py::TestQwen3_30B_A3B::test_nvfp4[dep4_latency_moe_cutlass]
- accuracy/test_llm_api_pytorch.py::TestLlama4ScoutInstruct::test_fp8[tp4-cuda_graph=True]
- accuracy/test_llm_api_pytorch.py::TestLlama4ScoutInstruct::test_fp4[tp4-cuda_graph=True]
- disaggregated/test_disaggregated.py::test_disaggregated_deepseek_v3_lite_fp8_ucx[DeepSeek-V3-Lite-fp8]
- disaggregated/test_disaggregated.py::test_disaggregated_deepseek_v3_lite_fp8_nixl[DeepSeek-V3-Lite-fp8]

7
tests/integration/test_lists/test-db/l0_dgx_h100.yml
View File

@ -132,18 +132,25 @@ l0_dgx_h100:
- cpp/test_multi_gpu.py::test_trt_gpt_real_decoder[llama-90]
- cpp/test_multi_gpu.py::TestDisagg::test_symmetric_executor[gpt-2proc-mpi_kvcache-90]
- cpp/test_multi_gpu.py::TestDisagg::test_symmetric_executor[gpt-2proc-ucx_kvcache-90]
- cpp/test_multi_gpu.py::TestDisagg::test_symmetric_executor[gpt-2proc-nixl_kvcache-90]
- cpp/test_multi_gpu.py::TestDisagg::test_symmetric_executor[llama-2proc-mpi_kvcache-90]
- cpp/test_multi_gpu.py::TestDisagg::test_symmetric_executor[llama-4proc-mpi_kvcache-90]
- cpp/test_multi_gpu.py::TestDisagg::test_symmetric_executor[llama-8proc-mpi_kvcache-90]
- cpp/test_multi_gpu.py::TestDisagg::test_symmetric_executor[llama-2proc-ucx_kvcache-90]
- cpp/test_multi_gpu.py::TestDisagg::test_symmetric_executor[llama-4proc-ucx_kvcache-90]
- cpp/test_multi_gpu.py::TestDisagg::test_symmetric_executor[llama-8proc-ucx_kvcache-90]
- cpp/test_multi_gpu.py::TestDisagg::test_symmetric_executor[llama-2proc-nixl_kvcache-90]
- cpp/test_multi_gpu.py::TestDisagg::test_symmetric_executor[llama-4proc-nixl_kvcache-90]
- cpp/test_multi_gpu.py::TestDisagg::test_symmetric_executor[llama-8proc-nixl_kvcache-90]
- cpp/test_multi_gpu.py::TestDisagg::test_asymmetric_executor[llama-4proc-mpi_kvcache-90]
- cpp/test_multi_gpu.py::TestDisagg::test_asymmetric_executor[llama-6proc-mpi_kvcache-90]
- cpp/test_multi_gpu.py::TestDisagg::test_asymmetric_executor[llama-8proc-mpi_kvcache-90]
- cpp/test_multi_gpu.py::TestDisagg::test_asymmetric_executor[llama-4proc-ucx_kvcache-90]
- cpp/test_multi_gpu.py::TestDisagg::test_asymmetric_executor[llama-6proc-ucx_kvcache-90]
- cpp/test_multi_gpu.py::TestDisagg::test_asymmetric_executor[llama-8proc-ucx_kvcache-90]
- cpp/test_multi_gpu.py::TestDisagg::test_asymmetric_executor[llama-4proc-nixl_kvcache-90]
- cpp/test_multi_gpu.py::TestDisagg::test_asymmetric_executor[llama-6proc-nixl_kvcache-90]
- cpp/test_multi_gpu.py::TestDisagg::test_asymmetric_executor[llama-8proc-nixl_kvcache-90]
- cpp/test_multi_gpu.py::TestDisagg::test_orchestrator_params[llama-mpi_kvcache-90]
- cpp/test_multi_gpu.py::TestDisagg::test_orchestrator_params[llama-ucx_kvcache-90]
- cpp/test_multi_gpu.py::TestDisagg::test_spawn_orchestrator[llama-ucx_kvcache-90]