Skip to content

GitLab

Unverified Commit 10a9ab7b authored by Kebe's avatar Kebe Committed by GitHub
Browse files

Fix error due to CustomAllreduce setup failure (#4815) 


Signed-off-by: default avatarKebe <mail@kebe7jun.com>
parent bb0fd749
Hide whitespace changes
Inline Side-by-side
Showing with 10 additions and 4 deletions
python/sglang/srt/distributed/parallel_state.py
View file @ 10a9ab7b
...@@ -264,10 +264,16 @@ class GroupCoordinator: ...@@ -264,10 +264,16 @@ class GroupCoordinator:
self.ca_comm: Optional[CustomAllreduce] = None self.ca_comm: Optional[CustomAllreduce] = None
if use_custom_allreduce and self.world_size > 1: if use_custom_allreduce and self.world_size > 1:
# Initialize a custom fast all-reduce implementation. # Initialize a custom fast all-reduce implementation.
self.ca_comm = CustomAllreduce( try:
group=self.cpu_group, self.ca_comm = CustomAllreduce(
device=self.device, group=self.cpu_group,
) device=self.device,
)
except Exception as e:
logger.warning(
f"Setup Custom allreduce failed with {e}. To silence this "
"warning, specify --disable-custom-all-reduce explicitly."
)
from sglang.srt.distributed.device_communicators.hpu_communicator import ( from sglang.srt.distributed.device_communicators.hpu_communicator import (
HpuCommunicator, HpuCommunicator,
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment