Skip to content

Handle constants without placeholder nodes in convert_constant_dim_order_pass #16841

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

Open
GregoryComer wants to merge 1 commit into
base: main
Choose a base branch
from
+10 −4
Open

Handle constants without placeholder nodes in convert_constant_dim_order_pass #16841

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
14 changes: 10 additions & 4 deletions exir/passes/convert_constant_dim_order_pass.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,7 +1,11 @@
import logging

import torch
from torch.export import ExportedProgram
from torch.export.graph_signature import InputKind

logger = logging.getLogger(__name__)


def _should_transform(tensor: torch.Tensor) -> bool:
"""
Expand All @@ -17,7 +21,7 @@ def _update_placeholder_meta(
exported_program: ExportedProgram,
target: str,
kind: InputKind,
):
) -> None:
input_spec = next(
(
spec
Expand All @@ -27,7 +31,8 @@ def _update_placeholder_meta(
None,
)
if input_spec is None:
raise RuntimeError(f"Missing input spec for lifted tensor {target}.")
logger.warning(f"Missing input spec for constant {target}")
Copy link
Contributor

@Gasoonjia Gasoonjia Jan 26, 2026

Choose a reason for hiding this comment

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

Do you know when the export program may not have input_spec? Doesn't every program have that?

Copy link
Member Author

@GregoryComer GregoryComer Jan 26, 2026

Choose a reason for hiding this comment

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

I can track down a better repro. This showed up with an internal CV model. My guess is that there is some pass that is modifying the constant and it hasn't been fully re-lifted at this point. I'll see if I can find an exact case. There are several broken internal tests for export currently.

return

placeholder_node = next(
(
Expand All @@ -37,8 +42,9 @@ def _update_placeholder_meta(
),
None,
)
if input_spec is None:
raise RuntimeError(f"Missing placeholder for {input_spec.arg.name}.")
if placeholder_node is None:
logger.warning(f"Missing placeholder node for constant {target}")
return

placeholder_node.meta["val"] = placeholder_node.meta["val"].contiguous()

Expand Down
Loading