minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

fix(metrics): double times add_latency for DECODE_BOOTSTRAP (#12209)

Browse Source
This commit is contained in:
Jimmy
2025-10-27 23:59:48 +08:00
committed by GitHub
parent 6c1c193308
commit 959d1ab84b
1 changed files with 0 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
python/sglang/srt/disaggregation/decode.py
-1
View File
@@ -521,7 +521,6 @@ class DecodePreallocQueue:
decode_req.kv_receiver.init( decode_req.kv_receiver.init(
page_indices, decode_req.metadata_buffer_index, state_indices page_indices, decode_req.metadata_buffer_index, state_indices
) )
decode_req.req.add_latency(RequestStage.DECODE_BOOTSTRAP)
preallocated_reqs.append(decode_req) preallocated_reqs.append(decode_req)
indices_to_remove.add(i) indices_to_remove.add(i)
decode_req.req.time_stats.decode_transfer_queue_entry_time = ( decode_req.req.time_stats.decode_transfer_queue_entry_time = (