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

[Lint]Style: Convert vllm-ascend/ to ruff format(Batch #10) (#6173)

Browse Source 
### What this PR does / why we need it?
**Scope of Changes**:
| File Path |
| :--- |
|`vllm_ascend/ops/layer_shard_linear.py`|
|`vllm_ascend/ops/linear.py`|
|`vllm_ascend/ops/linear_op.py`|
|`vllm_ascend/worker/worker.py`|
| ` vllm_ascend/patch/worker/patch_bert.py` |
| ` vllm_ascend/patch/worker/patch_deepseek.py` |
| ` vllm_ascend/patch/worker/patch_distributed.py` |
| ` vllm_ascend/patch/worker/patch_module.py` |
| ` vllm_ascend/patch/worker/patch_multimodal_merge.py` |
| ` vllm_ascend/patch/worker/patch_qwen3_next.py` |
| ` vllm_ascend/patch/worker/patch_qwen3_next_mtp.py` |
| ` vllm_ascend/patch/worker/patch_rejection_sampler.py` |
| ` vllm_ascend/patch/worker/patch_rope.py` |
| ` vllm_ascend/patch/worker/patch_triton.py` |
| ` vllm_ascend/patch/worker/patch_unquantized_gemm.py` |
| ` vllm_ascend/patch/worker/patch_v2_egale.py` |
|` vllm_ascend/worker/npu_input_batch.py`|
|` vllm_ascend/worker/v2/aclgraph_utils.py`|
|` vllm_ascend/worker/v2/attn_utils.py`|
|` vllm_ascend/worker/v2/model_runner.py`|
|` vllm_ascend/worker/v2/sample/gumbel.py`|
|` vllm_ascend/worker/v2/sample/penalties.py`|
|` vllm_ascend/worker/v2/sample/sampler.py`|
|` vllm_ascend/worker/v2/spec_decode/__init__.py`|
|` vllm_ascend/worker/v2/spec_decode/eagle.py`|
|` vllm_ascend/worker/v2/states.py`|
### Does this PR introduce _any_ user-facing change?

### How was this patch tested?

- vLLM version: v0.14.0
- vLLM main:
d68209402d

Signed-off-by: MrZ20 <2609716663@qq.com>
Signed-off-by: SILONG ZENG <2609716663@qq.com>
Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
Co-authored-by: wangxiyuan <wangxiyuan1007@gmail.com>
This commit is contained in:
SILONG ZENG
2026-02-06 15:35:06 +08:00
committed by GitHub
parent 65b7f716e6
commit 19b5d44ea8
33 changed files with 938 additions and 1243 deletions
Download Patch File Download Diff File Expand all files Collapse all files

44
vllm_ascend/patch/worker/patch_distributed.py
View File

@@ -15,29 +15,25 @@
# limitations under the License.
#
from typing import List, Optional, Union
import torch
import vllm
from torch.distributed import Backend
from vllm.distributed.parallel_state import (GroupCoordinator,
_get_unique_name, _register_group)
from vllm.distributed.parallel_state import GroupCoordinator, _get_unique_name, _register_group
from vllm_ascend.distributed.device_communicators.npu_communicator import \
NPUCommunicator
from vllm_ascend.distributed.device_communicators.npu_communicator import NPUCommunicator
from vllm_ascend.utils import create_hccl_pg_options
class GroupCoordinatorPatch(GroupCoordinator):
def __init__(
self,
group_ranks: list[list[int]],
local_rank: int,
torch_distributed_backend: Union[str, Backend],
torch_distributed_backend: str | Backend,
use_device_communicator: bool, # whether to use device communicator
use_message_queue_broadcaster: bool = False,
group_name: Optional[str] = None,
group_name: str | None = None,
):
group_name = group_name or "anonymous"
self.unique_name = _get_unique_name(group_name)
@@ -52,9 +48,8 @@ class GroupCoordinatorPatch(GroupCoordinator):
for ranks in group_ranks:
device_group = torch.distributed.new_group(
ranks,
backend=torch_distributed_backend,
pg_options=hccl_pg_options)
ranks, backend=torch_distributed_backend, pg_options=hccl_pg_options
)
# a group with `gloo` backend, to allow direct coordination between
# processes through the CPU.
@@ -83,22 +78,23 @@ class GroupCoordinatorPatch(GroupCoordinator):
unique_name=self.unique_name,
)
from vllm.distributed.device_communicators.shm_broadcast import \
MessageQueue
self.mq_broadcaster: Optional[MessageQueue] = None
from vllm.distributed.device_communicators.shm_broadcast import MessageQueue
self.mq_broadcaster: MessageQueue | None = None
if use_message_queue_broadcaster and self.world_size > 1:
self.mq_broadcaster = MessageQueue.create_from_process_group(
self.cpu_group, 1 << 22, 6)
self.mq_broadcaster = MessageQueue.create_from_process_group(self.cpu_group, 1 << 22, 6)
self.use_custom_op_call = False
self.use_cpu_custom_send_recv = False
def all_to_all(self,
input_: torch.Tensor,
scatter_dim: int = 0,
gather_dim: int = -1,
scatter_sizes: Optional[List[int]] = None,
gather_sizes: Optional[List[int]] = None) -> torch.Tensor:
def all_to_all(
self,
input_: torch.Tensor,
scatter_dim: int = 0,
gather_dim: int = -1,
scatter_sizes: list[int] | None = None,
gather_sizes: list[int] | None = None,
) -> torch.Tensor:
if self.world_size == 1:
return input_
assert -input_.dim() <= scatter_dim < input_.dim(), (
@@ -108,9 +104,7 @@ class GroupCoordinatorPatch(GroupCoordinator):
f"Invalid gather dim ({gather_dim}) for input tensor with shape {input_.size()}"
)
assert self.device_communicator is not None, "device_communicator should be initialized when world_size > 1"
return self.device_communicator.all_to_all(input_, scatter_dim,
gather_dim, scatter_sizes,
gather_sizes)
return self.device_communicator.all_to_all(input_, scatter_dim, gather_dim, scatter_sizes, gather_sizes)
def all_reduce(self, input_):
if self.world_size == 1: