xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-06-01 17:47:07 +08:00
Code Issues Packages Projects Releases Wiki Activity

minor

Browse Source 
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
This commit is contained in:
Woosuk Kwon 2024-12-26 11:13:41 -08:00
parent 52922193cd
commit ca4f9e69a8
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
vllm/v1/worker/gpu_block_table.py
View File

@ -36,9 +36,9 @@ class BlockTable:
) )
self.block_table_np = self.block_table_cpu.numpy() self.block_table_np = self.block_table_cpu.numpy()
self.use_uva = is_uva_available() # UVA requires pinned memory.
self.use_uva = is_uva_available() and pin_memory
if self.use_uva: if self.use_uva:
# Pinned memory is required to use UVA.
self.block_table_diff = torch.zeros((max_num_reqs, 2), self.block_table_diff = torch.zeros((max_num_reqs, 2),
dtype=torch.int32, dtype=torch.int32,
device="cpu", device="cpu",