Skip to content

Tiny add logs for expert location updater #7308

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
zhyncs merged 3 commits into from
Jun 28, 2025
Merged

Tiny add logs for expert location updater #7308

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
a8b1ec3
more
fzyzcjy Jun 18, 2025
57a4586
Merge branch 'main' into feat/update_weights_log
fzyzcjy Jun 18, 2025
2a31da9
Merge branch 'main' into feat/update_weights_log
fzyzcjy Jun 27, 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
16 changes: 16 additions & 0 deletions python/sglang/srt/model_executor/expert_location_updater.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -30,6 +30,9 @@
logger = logging.getLogger(__name__)


_LOG_INPUT = get_bool_env_var("SGLANG_EXPERT_LOCATION_UPDATER_LOG_INPUT")


class ExpertLocationUpdater:
def __init__(self):
self._first_execution = True
Expand Down Expand Up @@ -175,6 +178,19 @@ def update_expert_weights_single_layer(
assert isinstance(old_physical_to_logical_map, list)
assert isinstance(new_physical_to_logical_map, list)

if _LOG_INPUT:
logger.info(
"update_expert_weights_single_layer "
f"{[x.shape for x in routed_experts_weights]=} "
f"{[x.shape for x in temp_buffers]=} "
f"{old_physical_to_logical_map=} "
f"{new_physical_to_logical_map=} "
f"{num_local_physical_experts=} "
f"{num_gpu_per_node=} "
f"{rank=} "
f"{world_size=} "
)
Comment on lines +182 to +192
Copy link
Contributor

@gemini-code-assist gemini-code-assist bot Jun 18, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

Consider using logger.debug instead of logger.info for this type of verbose debugging output controlled by a specific environment variable. This allows finer-grained control over logging levels, ensuring this detailed information is only shown when explicitly requested for debugging. Additionally, the f-string formatting with = can produce very long lines, especially for lists of shapes or maps. While convenient, explicitly formatting the output for better readability might be beneficial.

Suggested change
logger.info(
"update_expert_weights_single_layer "
f"{[x.shape for x in routed_experts_weights]=} "
f"{[x.shape for x in temp_buffers]=} "
f"{old_physical_to_logical_map=} "
f"{new_physical_to_logical_map=} "
f"{num_local_physical_experts=} "
f"{num_gpu_per_node=} "
f"{rank=} "
f"{world_size=} "
)
logger.debug(
"update_expert_weights_single_layer "
f"{[x.shape for x in routed_experts_weights]=} "
f"{[x.shape for x in temp_buffers]=} "
f"{old_physical_to_logical_map=} "
f"{new_physical_to_logical_map=} "
f"{num_local_physical_experts=} "
f"{num_gpu_per_node=} "
f"{rank=} "
f"{world_size=} "
)


output_logs = [] if debug else None

num_physical_experts = len(old_physical_to_logical_map)
Expand Down
Loading