From f4489687862ab13cc10827f69e8ab4a95bba8628 Mon Sep 17 00:00:00 2001 From: JamesLim-sy Date: Thu, 6 Jun 2024 06:09:04 +0800 Subject: [PATCH 1/2] Addition of lack ignored_seq_groups in _schedule_chunked_prefill --- vllm/core/scheduler.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/vllm/core/scheduler.py b/vllm/core/scheduler.py index 399665082f838..eda8318fbed32 100644 --- a/vllm/core/scheduler.py +++ b/vllm/core/scheduler.py @@ -905,7 +905,7 @@ def _schedule_chunked_prefill(self): blocks_to_swap_out=running_scheduled.blocks_to_swap_out, blocks_to_copy=running_scheduled.blocks_to_copy + swapped_in.blocks_to_copy, - ignored_seq_groups=prefills.ignored_seq_groups, + ignored_seq_groups=prefills.ignored_seq_groups + swapped_in.infeasible_seq_groups, num_lookahead_slots=running_scheduled.num_lookahead_slots, running_queue_size=len(self.running), preempted=(len(running_scheduled.preempted) + From 3664f9c2889f2c41f16bd1dccf6c2d6ef1776c02 Mon Sep 17 00:00:00 2001 From: JamesLim-sy Date: Thu, 6 Jun 2024 23:46:42 +0800 Subject: [PATCH 2/2] revise code format --- vllm/core/scheduler.py | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/vllm/core/scheduler.py b/vllm/core/scheduler.py index eda8318fbed32..0159053b4dc6a 100644 --- a/vllm/core/scheduler.py +++ b/vllm/core/scheduler.py @@ -905,7 +905,8 @@ def _schedule_chunked_prefill(self): blocks_to_swap_out=running_scheduled.blocks_to_swap_out, blocks_to_copy=running_scheduled.blocks_to_copy + swapped_in.blocks_to_copy, - ignored_seq_groups=prefills.ignored_seq_groups + swapped_in.infeasible_seq_groups, + ignored_seq_groups=prefills.ignored_seq_groups + + swapped_in.infeasible_seq_groups, num_lookahead_slots=running_scheduled.num_lookahead_slots, running_queue_size=len(self.running), preempted=(len(running_scheduled.preempted) +