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 #8) (#6129)

Browse Source 
### What this PR does / why we need it?
**Scope of Changes**:
| File Path |
| :--- |
| vllm_ascend/ops/\_\_init\_\_.py |
| vllm_ascend/ops/activation.py |
| vllm_ascend/ops/flashcomm2_oshard_manager.py |
| vllm_ascend/ops/layernorm.py |
| vllm_ascend/ops/mla.py |
| vllm_ascend/ops/mm_encoder_attention.py |
| vllm_ascend/ops/register_custom_ops.py |
| vllm_ascend/ops/vocab_parallel_embedding.py |
| vllm_ascend/ops/weight_prefetch.py |
| vllm_ascend/spec_decode/\_\_init\_\_.py |
| vllm_ascend/spec_decode/eagle_proposer.py |
| vllm_ascend/spec_decode/interface.py |
| vllm_ascend/spec_decode/mtp_proposer.py |
| vllm_ascend/spec_decode/ngram_proposer.py |
| vllm_ascend/spec_decode/suffix_proposer.py |

### Does this PR introduce _any_ user-facing change?

### How was this patch tested?

- vLLM version: v0.13.0
- vLLM main:
d68209402d

Signed-off-by: MrZ20 <2609716663@qq.com>
Signed-off-by: SILONG ZENG <2609716663@qq.com>
This commit is contained in:
SILONG ZENG
2026-02-06 15:25:08 +08:00
committed by GitHub
parent 99aedaff63
commit 4fb3d5e1b2
17 changed files with 948 additions and 1147 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
vllm_ascend/ops/flashcomm2_oshard_manager.py
View File

@@ -1,11 +1,14 @@
from typing import Any, Dict, Optional
from typing import Any
from vllm.model_executor.models.utils import extract_layer_index
from vllm_ascend.distributed.parallel_state import get_shard_weight_group
from vllm_ascend.ops.layer_shard_linear import (
is_hidden_layer, post_process_after_loading_for_shard_weight_series,
reach_layer_for_shard_weight_series, register_layer_to_shard_weight_series)
is_hidden_layer,
post_process_after_loading_for_shard_weight_series,
reach_layer_for_shard_weight_series,
register_layer_to_shard_weight_series,
)
from vllm_ascend.utils import flashcomm2_enable, o_shard_enable
@@ -26,7 +29,7 @@ class Flashcomm2OShardManager:
"""
def __init__(self):
self._shard_layers: Dict[int, Any] = {}
self._shard_layers: dict[int, Any] = {}
def flashcomm2_oshard_enable(self):
return flashcomm2_enable() and o_shard_enable()
@@ -52,12 +55,10 @@ class Flashcomm2OShardManager:
self._shard_layers[layer_idx] = layer
register_layer_to_shard_weight_series(
series_name="o_proj",
group=get_shard_weight_group(),
layer=layer,
prefetch_step=prefetch_step)
series_name="o_proj", group=get_shard_weight_group(), layer=layer, prefetch_step=prefetch_step
)
def get_layer(self, layer_idx: int) -> Optional[Any]:
def get_layer(self, layer_idx: int) -> Any | None:
"""Safely retrieves a registered layer by its index.
Args: