Fix a bug in BatchTokenIDOut & Misc style and dependency updates (#7457)
This commit is contained in:
@@ -1374,7 +1374,14 @@ class Scheduler(
|
||||
)
|
||||
raise ValueError(msg)
|
||||
|
||||
if len(self.req_to_token_pool.free_slots) != self.req_to_token_pool.size:
|
||||
if self.disaggregation_mode == DisaggregationMode.DECODE:
|
||||
req_total_size = (
|
||||
self.req_to_token_pool.size + self.req_to_token_pool.pre_alloc_size
|
||||
)
|
||||
else:
|
||||
req_total_size = self.req_to_token_pool.size
|
||||
|
||||
if len(self.req_to_token_pool.free_slots) != req_total_size:
|
||||
msg = (
|
||||
"req_to_token_pool memory leak detected!"
|
||||
f"available_size={len(self.req_to_token_pool.free_slots)}, "
|
||||
|
||||
Reference in New Issue
Block a user