Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Minor] Small fix to make distributed init logic in worker looks cleaner #2905

Merged
merged 1 commit into from
Feb 18, 2024
Merged

[Minor] Small fix to make distributed init logic in worker looks cleaner #2905

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
[Minor] Small fix to make distributed init logic in worker looks cleaner
  • Loading branch information
@zhuohan123
zhuohan123 committed Feb 18, 2024
commit a1700ca3e1a0621a4798b9d9b361cc3d0cae1e71
6 changes: 4 additions & 2 deletions vllm/worker/worker.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -93,8 +93,6 @@ def init_model(self, cupy_port: Optional[int] = None) -> None:
# Initialize the distributed environment.
init_distributed_environment(self.parallel_config, self.rank,
cupy_port, self.distributed_init_method)
if not self.parallel_config.disable_custom_all_reduce:
init_custom_ar()
# Initialize the model.
set_random_seed(self.model_config.seed)

Expand Down Expand Up @@ -288,6 +286,10 @@ def init_distributed_environment(
ensure_model_parallel_initialized(parallel_config.tensor_parallel_size,
parallel_config.pipeline_parallel_size)

# Initialize a custom fast all-reduce implementation.
if not parallel_config.disable_custom_all_reduce:
init_custom_ar()


def _check_if_gpu_supports_dtype(torch_dtype: torch.dtype):
# Check if the GPU supports the dtype.
Expand Down
Loading