kanshan/TensorRT-LLMs
1
0
Fork 0
mirror of https://github.com/NVIDIA/TensorRT-LLM.git synced 2026-02-21 18:25:20 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

[None][test] Clean cache for certain easily hang cases (#8619)

Browse Source 
Signed-off-by: Ivy Zhang <25222398+crazydemo@users.noreply.github.com>
Co-authored-by: Larry Xu <197874197+LarryXFly@users.noreply.github.com>
This commit is contained in:
Ivy Zhang 2025-10-24 20:17:32 +08:00 committed by GitHub
parent 7749ec406b
commit 1859b55d22
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
6 changed files with 10 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
tests/integration/defs/conftest.py
View File

@ -15,6 +15,7 @@
# -*- coding: utf-8 -*-
import datetime
import gc
import os
import platform
import re
@ -2562,4 +2563,5 @@ def torch_empty_cache() -> None:
Manually empty the torch CUDA cache before each test, to reduce risk of OOM errors.
"""
if torch.cuda.is_available():
gc.collect()
torch.cuda.empty_cache()

4
tests/integration/defs/test_e2e.py
View File

@ -2262,7 +2262,7 @@ def test_ptp_quickstart_advanced_deepseek_r1_w4afp8_8gpus(
@pytest.mark.skip_less_device_memory(80000)
@pytest.mark.parametrize("model_name,model_path,gpu_count", [
("Llama3.1-70B-BF16", "llama-3.1-model/Meta-Llama-3.1-70B", 2),
("Llama3.1-70B-BF16", "llama-3.1-model/Meta-Llama-3.1-70B", 8),
("Mixtral-8x7B-BF16", "Mixtral-8x7B-v0.1", 8),
pytest.param('Llama3.1-70B-FP8',
'llama-3.1-model/Llama-3.1-70B-Instruct-FP8',
@ -2293,7 +2293,7 @@ def test_ptp_quickstart_advanced_multi_gpus(llm_root, llm_venv, model_name,
pytest.skip(f"Not enough GPUs for {model_name}")
example_root = Path(os.path.join(llm_root, "examples", "llm-api"))
mapping = {
"Llama3.1-70B-BF16": 91.0,
"Llama3.1-70B-BF16": 24.6,
"Mixtral-8x7B-BF16": 16.5,
"Llama3.1-70B-FP8": 58.5,
"Llama3.1-405B-FP8": 63.2,

2
tests/integration/test_lists/qa/llm_function_core.txt
View File

@ -634,7 +634,7 @@ test_e2e.py::test_ptp_quickstart_advanced[Llama3.1-8B-NVFP4-nvfp4-quantized/Meta
test_e2e.py::test_ptp_quickstart_advanced[Llama3.2-11B-BF16-llama-3.2-models/Llama-3.2-11B-Vision]
test_e2e.py::test_ptp_quickstart_advanced[Qwen3-30B-A3B-Qwen3/Qwen3-30B-A3B]
test_e2e.py::test_ptp_quickstart_advanced_ngram[Llama-3.1-8B-Instruct-llama-3.1-model/Llama-3.1-8B-Instruct]
test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Llama3.1-70B-BF16-llama-3.1-model/Meta-Llama-3.1-70B-2]
test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Llama3.1-70B-BF16-llama-3.1-model/Meta-Llama-3.1-70B-8]
test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Llama3.1-70B-FP8-llama-3.1-model/Llama-3.1-70B-Instruct-FP8-2]
test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Llama3.1-405B-FP8-llama-3.1-model/Llama-3.1-405B-Instruct-FP8-8]
test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Mixtral-8x7B-BF16-Mixtral-8x7B-v0.1-8]

2
tests/integration/test_lists/qa/llm_function_core_sanity.txt
View File

@ -210,7 +210,7 @@ test_e2e.py::test_openai_consistent_chat
test_e2e.py::test_openai_multi_chat_example
test_e2e.py::test_ptp_quickstart
test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Llama3.1-405B-FP8-llama-3.1-model/Llama-3.1-405B-Instruct-FP8-8]
test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Llama3.1-70B-BF16-llama-3.1-model/Meta-Llama-3.1-70B-2]
test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Llama3.1-70B-BF16-llama-3.1-model/Meta-Llama-3.1-70B-8]
test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Llama3.1-70B-FP8-llama-3.1-model/Llama-3.1-70B-Instruct-FP8-2]
test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Mixtral-8x7B-BF16-Mixtral-8x7B-v0.1-8]
test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Mixtral-8x7B-NVFP4-nvfp4-quantized/Mixtral-8x7B-Instruct-v0.1-8]

2
tests/integration/test_lists/qa/llm_function_nim.txt
View File

@ -408,7 +408,7 @@ test_e2e.py::test_ptp_quickstart_advanced[Llama3.1-8B-BF16-llama-3.1-model/Meta-
test_e2e.py::test_ptp_quickstart_advanced[Llama3.1-8B-FP8-llama-3.1-model/Llama-3.1-8B-Instruct-FP8]
test_e2e.py::test_ptp_quickstart_advanced[Llama3.1-8B-NVFP4-nvfp4-quantized/Meta-Llama-3.1-8B]
test_e2e.py::test_ptp_quickstart_advanced[Qwen3-30B-A3B-Qwen3/Qwen3-30B-A3B]
test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Llama3.1-70B-BF16-llama-3.1-model/Meta-Llama-3.1-70B-2]
test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Llama3.1-70B-BF16-llama-3.1-model/Meta-Llama-3.1-70B-8]
test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Llama3.1-70B-FP8-llama-3.1-model/Llama-3.1-70B-Instruct-FP8-2]
test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Llama3.1-405B-FP8-llama-3.1-model/Llama-3.1-405B-Instruct-FP8-8]
test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[Mixtral-8x7B-BF16-Mixtral-8x7B-v0.1-8]

4
tests/integration/test_lists/waives.txt
View File

@ -313,7 +313,6 @@ full:H100/accuracy/test_llm_api_pytorch.py::TestLlama4MaverickInstruct::test_fp8
full:H100/accuracy/test_llm_api_pytorch.py::TestLlama4MaverickInstruct::test_fp8[tp8-cuda_graph=True] SKIP (https://nvbugs/5512734)
full:H100/accuracy/test_llm_api_pytorch.py::TestLlama4MaverickInstruct::test_fp8_eagle3[tp8-torch_compile=True] SKIP (https://nvbugs/5483534)
full:A100/test_e2e.py::test_ptp_quickstart_multimodal[NVILA-8B-FP16-vila/NVILA-8B-video-False] SKIP (https://nvbugs/5453725)
test_e2e.py::test_ptp_scaffolding[DeepSeek-R1-Distill-Qwen-7B-DeepSeek-R1/DeepSeek-R1-Distill-Qwen-7B] SKIP (https://nvbugs/5517260)
test_e2e.py::test_ptp_quickstart_multimodal[NVILA-8B-FP16-vila/NVILA-8B-image-False] SKIP (https://nvbugs/5509024)
test_e2e.py::test_ptp_quickstart_multimodal[qwen2.5-vl-7b-instruct-Qwen2.5-VL-7B-Instruct-video-False] SKIP (https://nvbugs/5509024)
test_e2e.py::test_ptp_quickstart_multimodal[qwen2.5-vl-7b-instruct-Qwen2.5-VL-7B-Instruct-video-True] SKIP (https://nvbugs/5509024)
@ -347,3 +346,6 @@ triton_server/test_triton_llm.py::test_llmapi_backend[1-0-enableDecoupleMode-ten
cpp/test_e2e.py::test_benchmarks[gpt-80] SKIP (https://nvbugs/5601670)
disaggregated/test_disaggregated.py::test_disaggregated_deepseek_v3_lite_bf16_empty_batch[DeepSeek-V3-Lite-bf16] SKIP (https://nvbugs/5601682)
disaggregated/test_disaggregated.py::test_disaggregated_benchmark_on_diff_backends[llama-v3-8b-hf] SKIP (https://nvbugs/5587574)
full:H20-3e/accuracy/test_llm_api_pytorch.py::TestNemotronUltra::test_auto_dtype[tp8ep4-cuda_graph=True] SKIP (slow I/O)
full:H20-3e/accuracy/test_llm_api_pytorch.py::TestKimiK2::test_fp8_blockscale[latency] SKIP (slow I/O)
full:H20-3e/test_e2e.py::test_ptp_quickstart_advanced_multi_gpus[DeepSeek-V3-671B-FP8-DeepSeek-V3-0324-8] SKIP (slow I/O)