[PD] Fix failure abort (#6535)

This commit is contained in:
Byron Hsu
2025-05-22 20:32:03 -07:00
committed by GitHub
parent a6970a17f3
commit 0a4fc73b48
6 changed files with 141 additions and 92 deletions

View File

@@ -582,6 +582,8 @@ class Scheduler(
gloo_group=self.attn_tp_cpu_group,
req_to_metadata_buffer_idx_allocator=req_to_metadata_buffer_idx_allocator,
metadata_buffers=metadata_buffers,
scheduler=self,
tree_cache=self.tree_cache,
)
# The decode requests pending for pre-allocation