xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-20 07:25:01 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Bugfix] Free cross attention block table for preempted-for-recompute sequence group. (#10013)

Browse Source 
Signed-off-by: Kathy Yu <feiyangyu@google.com>
This commit is contained in:
Kathy Yu 2025-01-02 10:28:09 -08:00 committed by GitHub
parent 84c35c374a
commit 2f385183f3
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 1 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
vllm/core/scheduler.py
View File

@ -1579,6 +1579,7 @@ class Scheduler:
seq.status = SequenceStatus.WAITING seq.status = SequenceStatus.WAITING
self.free_seq(seq) self.free_seq(seq)
seq.reset_state_for_recompute() seq.reset_state_for_recompute()
self._free_seq_group_cross_attn_blocks(seq_group)
def _preempt_by_swap( def _preempt_by_swap(
self, self,