apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-12 20:11:20 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[Feature] Support block scheduler v1 for FD (#2928)

Browse Source 
* Support FD block scheduler v1

* Support FD block scheduler v1

* Support FD block scheduler v1

* Fix according to copilot review

* Fix according to review

* Remove is_dummy

* Fix bug when real_bsz=1

* Fix infer first token cost time

---------

Co-authored-by: Jiang-Jia-Jun <163579578+Jiang-Jia-Jun@users.noreply.github.com>
This commit is contained in:
chenjian
2025-07-23 20:31:31 +08:00
committed by GitHub
parent ca0f71bd39
commit 85a78d695d
16 changed files with 898 additions and 40 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
fastdeploy/output/token_processor.py
View File

@@ -25,6 +25,7 @@ from concurrent.futures import ThreadPoolExecutor
import numpy as np
from fastdeploy import envs
from fastdeploy.engine.request import CompletionOutput, RequestMetrics, RequestOutput
from fastdeploy.inter_communicator import IPCSignal
from fastdeploy.metrics.metrics import main_process_metrics
@@ -269,9 +270,12 @@ class TokenProcessor:
else:
time.sleep(0.002)
else:
self.resource_manager.stop_flags[index] = True
self.resource_manager.tasks_list[index] = None
self.resource_manager._recycle_block_tables(task)
if envs.ENABLE_V1_KVCACHE_SCHEDULER:
self.resource_manager.finish_requests_async(task_id)
else:
self.resource_manager.stop_flags[index] = True
self.resource_manager.tasks_list[index] = None
self.resource_manager._recycle_block_tables(task)
if task_id in self.tokens_counter:
del self.tokens_counter[task_id]
@@ -508,6 +512,7 @@ class TokenProcessor:
self.tokens_counter[task_id] += 1
if token_id != RECOVERY_STOP_SIGNAL:
result.outputs.token_ids.append(token_id)
task.output_token_ids.append(token_id)
if token_id in task.eos_token_ids or is_prefill or recovery_stop:
result.finished = True
if recovery_stop: