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

Fix some ci issue and refactor modelrunner (#2445)

Browse Source 
### What this PR does / why we need it?
Fix some ci issue and refactor modelrunner

### Does this PR introduce _any_ user-facing change?
N/A

### How was this patch tested?
CI passed with existing test.

- vLLM version: v0.10.0
- vLLM main:
4d9c61993a

---------

Signed-off-by: wangli <wangli858794774@gmail.com>
Signed-off-by: MengqingCao <cmq0113@163.com>
Signed-off-by: weiguihua2 <weiguihua2@huawei.com>
Co-authored-by: wangli <wangli858794774@gmail.com>
Co-authored-by: weiguihua2 <weiguihua2@huawei.com>
This commit is contained in:
Mengqing Cao
2025-08-20 09:01:04 +08:00
committed by GitHub
parent 955411611c
commit 1327f9be1c
28 changed files with 1612 additions and 1020 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
vllm_ascend/worker/eagle_proposer_v1.py
View File

@@ -16,6 +16,7 @@ from vllm.v1.sample.metadata import SamplingMetadata
from vllm_ascend.ascend_forward_context import set_ascend_forward_context
from vllm_ascend.attention.attention_mask import AttentionMaskBuilder
from vllm_ascend.attention.attention_v1 import AscendAttentionState
from vllm_ascend.attention.utils import AscendCommonAttentionMetadata
PADDING_SLOT_ID = -1
@@ -125,12 +126,27 @@ class EagleProposer:
query_lens = cu_num_tokens[1:] - cu_num_tokens[:-1]
max_query_len = query_lens.max().item()
# FIXME(woosuk): The below two ops cause synchronization. Optimize.
attn_metadata = self.runner.attn_metadata_builder.build(
common_attn_metadata = AscendCommonAttentionMetadata(
query_start_loc=self.runner.query_start_loc[:batch_size + 1],
query_start_loc_cpu=self.runner.query_start_loc_cpu[:batch_size +
1],
seq_lens_cpu=self.runner.seq_lens_cpu,
max_query_len=max_query_len,
num_reqs=batch_size,
num_actual_tokens=num_tokens,
max_query_len=max_query_len,
actual_seq_lengths_q=self.runner.actual_seq_lengths_q,
block_table_tensor=self.runner.input_batch.block_table[0].
get_device_tensor(),
slot_mapping_cpu=target_slot_mapping,
positions=target_positions,
attn_mask=self.runner.attn_mask,
spec_attn_mask=self.runner.spec_attn_mask,
attn_state=self.runner.attn_state,
decode_token_per_req=self.runner.decode_token_per_req,
)
# FIXME(woosuk): The below two ops cause synchronization. Optimize.
attn_metadata = self.runner.attn_metadata_builder.build(
common_attn_metadata, self.runner.model)
if self.use_cuda_graph and \
num_tokens <= self.cudagraph_batch_sizes[-1]:
num_input_tokens = self.vllm_config.pad_for_cudagraph(num_tokens)