DeepSeek/DeepEP
1
0
Fork 0
mirror of https://github.com/deepseek-ai/DeepEP synced 2025-06-26 18:28:11 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Feature: LL nvlink p2p (#173)

Browse Source
This commit is contained in:
cywork121
2025-05-23 10:37:45 +08:00
committed by GitHub
parent d5ca4495c0
commit 68ae8b3d07
2 changed files with 34 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
deep_ep/buffer.py
View File

@@ -68,7 +68,8 @@ class Buffer:
if self.runtime.get_num_rdma_ranks() > 1 or low_latency_mode:
# Enable IBGDA
assert num_qps_per_rank > 0
os.environ['NVSHMEM_DISABLE_P2P'] = '1'
if not os.getenv("NVSHMEM_DISABLE_P2P"):
os.environ['NVSHMEM_DISABLE_P2P'] = '1'
os.environ['NVSHMEM_IB_ENABLE_IBGDA'] = '1'
os.environ['NVSHMEM_IBGDA_NIC_HANDLER'] = 'gpu'
os.environ['NVSHMEM_IBGDA_NUM_RC_PER_PE'] = f'{num_qps_per_rank}'