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

[Feature] Refactor PCP &DCP related code (#5214)

Browse Source 
### What this PR does / why we need it?
Refactor pcp& dcp related code. we use pcp_manager class to Unifiy
Manage pcp & dcp . as we do this , many code can be deleted from
model_runner, and can avoid break pcp & dcp by other developments.
RFC：https://github.com/vllm-project/vllm-ascend/issues/5449
### Does this PR introduce _any_ user-facing change?
NO

### How was this patch tested?

- vLLM version: release/v0.13.0
- vLLM main:
ad32e3e19c

---------

Signed-off-by: zhenwenqi2024 <zhenwenqi_2022@qq.com>
Co-authored-by: zzzzwwjj <34335947+zzzzwwjj@users.noreply.github.com>
This commit is contained in:
zhenwenqi2024
2025-12-31 09:29:57 +08:00
committed by GitHub
parent 46862ce1af
commit 5d9fde9819
7 changed files with 1156 additions and 1047 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
vllm_ascend/spec_decode/eagle_proposer.py
View File

@@ -279,9 +279,9 @@ class EagleProposer(VllmEagleProposer):
req_scheduled_tokens = scheduler_output.num_scheduled_tokens
if self.pcp_size > 1:
long_seq_metadata = self.runner.long_seq_metadata
input_ids_pcp_full = self.runner.input_ids_pcp_full
query_start_loc_pcp_full = self.runner.query_start_loc_pcp_full
query_start_loc_pcp_full_cpu = self.runner.query_start_loc_pcp_full_cpu
input_ids_pcp_full = self.runner.pcp_manager.input_ids_pcp_full.gpu
query_start_loc_pcp_full = self.runner.pcp_manager.query_start_loc_pcp_full.gpu
query_start_loc_pcp_full_cpu = self.runner.pcp_manager.query_start_loc_pcp_full.cpu
num_reqs = self.runner.input_batch.num_reqs
ori_query_lens = query_start_loc_pcp_full_cpu[1:num_reqs+1] - \
query_start_loc_pcp_full_cpu[:num_reqs]