apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-05 08:37:06 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[XPU]Fixed the issue of performance degradation caused by enabling ENABLE_V1_KVCACHE_SCHEDULER (#3897)

Browse Source 
* fix bug

* fix bug

* update

* update

* update
This commit is contained in:
yinwei
2025-09-08 10:34:46 +08:00
committed by GitHub
parent b649494655
commit 7833f2f6cb
4 changed files with 20 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
fastdeploy/engine/sched/resource_manager_v1.py
View File

@@ -363,7 +363,9 @@ class ResourceManagerV1(ResourceManager):
while self.waiting and token_budget > 0:
if len(self.running) == self.max_num_seqs:
break
if self.config.model_config.enable_mm and self.exist_prefill(scheduled_reqs):
if (self.config.model_config.enable_mm or paddle.is_compiled_with_xpu()) and self.exist_prefill(
scheduled_reqs
):
break
request = self.waiting[0]
if request.status == RequestStatus.WAITING: