Skip to content

[Bug Fix] fix vl V1 schedule bug #3284

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
Aug 11, 2025
Merged

[Bug Fix] fix vl V1 schedule bug #3284

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
78b82ee
[Bug Fix] fix vl V1 schedule bug
ming1753 Aug 8, 2025
ae6dc24
Merge branch 'release/2.1' into vl_v1
Jiang-Jia-Jun Aug 11, 2025
2ba61d0
Merge branch 'release/2.1' into vl_v1
XieYunshen Aug 11, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 0 additions & 1 deletion fastdeploy/engine/sched/resource_manager_v1.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -195,7 +195,6 @@ def _get_num_new_tokens(self, request, token_budget):
)
request.num_image_end = img_num_per_boundary[new_boundary_idx]

request.num_image_end = img_num_per_boundary[new_boundary_idx]
request.image_type_ids_start = np.sum(grid_thw[: request.num_image_start, 0])
request.image_type_ids_end = np.sum(grid_thw[: request.num_image_end, 0])
request.image_start = np.sum(np.prod(grid_thw[: request.num_image_start], axis=1))
Expand Down
6 changes: 4 additions & 2 deletions fastdeploy/worker/worker_process.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,6 +24,7 @@
import paddle.distributed as dist
from paddle.distributed import fleet

from fastdeploy import envs
from fastdeploy.config import (
CacheConfig,
DecodingConfig,
Expand Down Expand Up @@ -289,8 +290,9 @@ def event_loop_normal(self) -> None:
if self.local_rank % mp_num_per_node == 0:
if self.task_queue.num_tasks() > 0:
# VL only support 1 batch to prefill

if not self.fd_config.model_config.enable_mm or not self.worker.exist_prefill():
if envs.ENABLE_V1_KVCACHE_SCHEDULER or not (
self.fd_config.model_config.enable_mm and self.worker.exist_prefill()
):
if self.nnode > 1 and self.parallel_config.tensor_parallel_size > self.max_chips_per_node:
self.task_queue.read_finish_flag.set(1)
else:
Expand Down
Loading