[Bugfix][Core] Exempt exactly the blocks an async KV load writes from zeroing (#59504)

Signed-off-by: Yifan Qiao <yifanqiao@inferact.ai>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Codex <noreply@openai.com>
This commit is contained in:
Yifan Qiao
2026-10-03 06:26:12 +00:00
committed by GitHub
co-authored by Claude Opus 5.5 Codex
parent 1a001d5842
commit ca0df1f122
17 changed files with 162 additions and 72 deletions
@@ -192,6 +192,7 @@ def get_fake_allocate_slots_fn(original_allocate_slots_fn: Callable):
full_sequence_must_fit: bool = False,
reserved_blocks: int = 0,
has_scheduled_reqs: bool = True,
skip_zeroing_group_ids: tuple[int, ...] = (),
):
ret = original_allocate_slots_fn(
self,
@@ -206,6 +207,7 @@ def get_fake_allocate_slots_fn(original_allocate_slots_fn: Callable):
full_sequence_must_fit,
reserved_blocks,
has_scheduled_reqs,
skip_zeroing_group_ids,
)
if cur_step_action is not None:
cur_block_ids = self.coordinator.single_type_managers[0].req_to_blocks[