Skip to content

DTensor: add more foreach ops to supported sharding prop list #132066

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

Closed
bdhirsh wants to merge 5 commits into from
Closed

DTensor: add more foreach ops to supported sharding prop list #132066

Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
eaad72f
DTensor: use pytrees to convert DTensors into specs
bdhirsh Jul 29, 2024
10615ec
Update on "DTensor: use pytrees to convert DTensors into specs"
bdhirsh Jul 29, 2024
e030707
Update on "DTensor: use pytrees to convert DTensors into specs"
bdhirsh Jul 29, 2024
1303b58
Update on "DTensor: use pytrees to convert DTensors into specs"
bdhirsh Jul 29, 2024
5036dd9
Update on "DTensor: use pytrees to convert DTensors into specs"
bdhirsh Aug 1, 2024
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
8 changes: 8 additions & 0 deletions torch/distributed/_tensor/ops/_pointwise_ops.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -545,10 +545,18 @@ def linear_pointwise_strategy(mesh: DeviceMesh, op_schema: OpSchema) -> Strategy
aten._foreach_clamp_max_.Scalar,
aten._foreach_clamp_min_.Scalar,
aten._foreach_div_.List,
Copy link
Contributor

@wz337 wz337 Jul 29, 2024

Choose a reason for hiding this comment

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

Should we also add aten. foreach_div.Scalar and aten. _foreach_div.Scalar in this PR as well?

Copy link
Contributor Author

@bdhirsh bdhirsh Jul 29, 2024

Choose a reason for hiding this comment

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

yep good call

aten._foreach_div_.Scalar,
aten._foreach_div_.ScalarList,
aten._foreach_div_.Tensor,
aten._foreach_div.List,
aten._foreach_div.Scalar,
aten._foreach_div.ScalarList,
aten._foreach_div.Tensor,
aten._foreach_lerp_.Scalar,
aten._foreach_maximum_.List,
aten._foreach_mul.Scalar,
aten._foreach_mul.ScalarList,
aten._foreach_mul.Tensor,
aten._foreach_mul.List,
aten._foreach_mul_.Scalar,
aten._foreach_mul_.ScalarList,
Expand Down