Skip to content

Tag mutated buffer for AOTI cuda partitioner #14783

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 1 commit into from
Oct 3, 2025
Merged

Tag mutated buffer for AOTI cuda partitioner #14783

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
3 changes: 2 additions & 1 deletion backends/cuda/cuda_partitioner.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,7 +15,7 @@
Partitioner,
PartitionResult,
)
from executorch.exir.backend.utils import tag_constant_data
from executorch.exir.backend.utils import tag_constant_data, tag_mutated_buffer
from torch.export.exported_program import ExportedProgram


Expand Down Expand Up @@ -54,6 +54,7 @@ def partition(self, exported_program: ExportedProgram) -> PartitionResult:
partition_tags[tag] = self.delegation_spec

tag_constant_data(exported_program)
tag_mutated_buffer(exported_program)
Copy link
Contributor

@mergennachin mergennachin Oct 3, 2025 •
edited
Loading

Choose a reason for hiding this comment

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

Make inline comment, something like this

  # Tag mutated buffers (e.g., KV caches) to avoid unnecessary copies.
  # Buffers that are mutated in-place within the delegated subgraph
  # don't need to be returned as outputs.


return PartitionResult(
tagged_exported_program=exported_program, partition_tags=partition_tags
Expand Down
Loading