900086fdc6
### What this PR does / why we need it?
Part of https://github.com/vllm-project/vllm-ascend/pull/3106
Fix Hybrid kvcache sharing bug in same attention type
Change the `shared_by` logic so that the same attention spec could share
the same buffer instead of allocating more hbm.
After this pr, kvcache memory saved 50% in qwen3-next compared with
before (`self_attn:linear_attn=1:3` in an `attn_group`), and
`gpu_memory_utilization` could increase to `0.8` on Qwen3-Next when
running on A2 64G/card with tp4
<img width="2833" height="1540" alt="image"
src="https://github.com/user-attachments/assets/2a91fa99-fb0f-447c-9e8b-acd587890fbe"
/>
### Does this PR introduce _any_ user-facing change?
### How was this patch tested?
Test pass with the latest e2e test case on qwen3-next
- vLLM version: v0.11.0rc3
- vLLM main:
c9461e05a4
---------
Signed-off-by: MengqingCao <cmq0113@163.com>
39 lines
1.4 KiB
Python
39 lines
1.4 KiB
Python
#
|
|
# Copyright (c) 2025 Huawei Technologies Co., Ltd. All Rights Reserved.
|
|
# Copyright 2023 The vLLM team.
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
# This file is a part of the vllm-ascend project.
|
|
# Adapted from vllm/tests/basic_correctness/test_basic_correctness.py
|
|
#
|
|
"""Compare the short outputs of HF and vLLM when using greedy sampling.
|
|
|
|
Run `pytest tests/e2e/multicard/test_qwen3_next.py`.
|
|
"""
|
|
|
|
from tests.e2e.conftest import VllmRunner
|
|
|
|
|
|
def test_models_distributed_Qwen3_NEXT_TP4():
|
|
example_prompts = [
|
|
"Hello, my name is",
|
|
] * 4
|
|
max_tokens = 5
|
|
with VllmRunner("Qwen/Qwen3-Next-80B-A3B-Instruct",
|
|
tensor_parallel_size=4,
|
|
max_model_len=4096,
|
|
gpu_memory_utilization=0.8,
|
|
distributed_executor_backend="mp",
|
|
enforce_eager=True) as vllm_model:
|
|
vllm_model.generate_greedy(example_prompts, max_tokens)
|