biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[DCP] Support dcp kv_cache interleave size > 1 (#26696)

Browse Source 
Signed-off-by: zhangsicheng5 <zhangsicheng5@huawei.com>
Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com>
Signed-off-by: Qiu <qiuchunshuo@huawei.com>
Co-authored-by: QiuChunshuo <qiuchunshuo@huawei.com>
This commit is contained in:
zhangsicheng5
2025-11-09 03:45:27 +08:00
committed by GitHub
parent 47604137a2
commit 2108a571d7
12 changed files with 202 additions and 79 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
tests/v1/worker/test_gpu_model_runner.py
View File

@@ -951,6 +951,7 @@ def test_hybrid_block_table_initialization():
max_num_reqs = 10
max_num_blocks_per_req = 20
max_num_batched_tokens = 512
dcp_kv_cache_interleave_size = 8
block_table = BlockTable(
block_size=block_size,
@@ -960,6 +961,7 @@ def test_hybrid_block_table_initialization():
pin_memory=False,
device=torch.device(DEVICE),
kernel_block_size=kernel_block_sizes[0],
dcp_kv_cache_interleave_size=dcp_kv_cache_interleave_size,
)
# Verify hybrid block configuration