EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

[BugFix] fix metadata execute error: integer modulo by zero (#6521)

Browse Source 
### What this PR does / why we need it?
fix metadata execute error: integer modulo by zero 

- vLLM version: v0.15.0
- vLLM main: https://github.com/vllm-project/vllm/commit/v0.15.0

Signed-off-by: xleoken <xleoken@163.com>
This commit is contained in:
xleoken
2026-03-10 09:58:06 +08:00
committed by GitHub
parent f6db47f103
commit 146b9d2a83
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
vllm_ascend/distributed/kv_transfer/kv_pool/cpu_offload/cpu_kv_cache_manager.py
View File

@@ -65,7 +65,7 @@ class CPUKVCacheManager:
self.num_cpu_blocks = num_cpu_blocks
self.caching_hash_fn = sha256 if caching_hash_algo == "sha256" else hash
self.use_eagle = use_eagle
self.block_pool = BlockPool(self.num_cpu_blocks, True, enable_kv_cache_events)
self.block_pool = BlockPool(self.num_cpu_blocks, True, self.block_size, enable_kv_cache_events)
self.single_type_manager = get_manager_for_kv_cache_spec(
kv_cache_spec=kv_cache_spec,
block_pool=self.block_pool,