kanshan/TensorRT-LLMs
1
0
Fork 0
mirror of https://github.com/NVIDIA/TensorRT-LLM.git synced 2026-02-04 18:21:52 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

[None][chore] Correct sorting order for attention DP scheduling to prioritize non-relaxed requests (#11106)

Browse Source 
Signed-off-by: Lance Liao <108499334+lancelly@users.noreply.github.com>
This commit is contained in:
Liao Lanyu 2026-01-30 16:06:48 +08:00 committed by GitHub
parent 322471cdd7
commit f2dd0ee128
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
tensorrt_llm/_torch/pyexecutor/executor_request_queue.py
View File

@ -445,7 +445,7 @@ class ExecutorRequestQueue:
return True
return scheduling_params.attention_dp_relax
new_requests = sorted(new_requests, key=get_relax_value, reverse=True)
new_requests = sorted(new_requests, key=get_relax_value)
# Try to put the requests to the target dp rank until the max_num_active_requests is reached
remaining_unscheduled = []