-
Notifications
You must be signed in to change notification settings - Fork 722
Make make_tensor in broadcast utilities public and rename free_broadcast_tensor #2785
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.
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
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/2785
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit af92d4e with merge base 9c38cf7 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
|
This pull request was exported from Phabricator. Differential Revision: D55577026 |
…ast_tensor (#2785) Summary: This diff does a couple of things: - Makes `make_tensor` a public function so that we can create temporary intermediate tensors in operators that need to do so. (Such as NMS that is implemented above in this stack) - Renames `free_broadcast_tensor` to a more generic name `free_tensor` Differential Revision: D55577026
…ast_tensor (#2785) Summary: This diff does a couple of things: - Makes `make_tensor` a public function so that we can create temporary intermediate tensors in operators that need to do so. (Such as NMS that is implemented above in this stack) - Renames `free_broadcast_tensor` to a more generic name `free_tensor` Differential Revision: D55577026
cb1f418 to
f70550f
Compare
|
This pull request was exported from Phabricator. Differential Revision: D55577026 |
…ast_tensor (#2785) Summary: This diff does a couple of things: - Makes `make_tensor` a public function so that we can create temporary intermediate tensors in operators that need to do so. (Such as NMS that is implemented above in this stack) - Renames `free_broadcast_tensor` to a more generic name `free_tensor` Differential Revision: D55577026
| } | ||
|
|
||
| namespace { | ||
|
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It's interesting that malloc is used in this function but I don't see how the malloced pointer is freed. @SS-JIA @manuelcandales , could you provide more context that I miss? cc @dbort
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
There's a corresponding function called free_tensor that when passed in the tensor created via this function free's it up.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@tarun292 These functions are not being used in the portable kernels. I believe you introduced these utils back in January 2023, and maybe they were being used at that time, but not anymore. They should be moved out of the portable lib utils.
…ast_tensor (#2785) Summary: This diff does a couple of things: - Makes `make_tensor` a public function so that we can create temporary intermediate tensors in operators that need to do so. (Such as NMS that is implemented above in this stack) - Renames `free_broadcast_tensor` to a more generic name `free_tensor` Differential Revision: D55577026
Differential Revision: D55577025
…ast_tensor (#2785) Summary: Pull Request resolved: #2785 This diff does a couple of things: - Makes `make_tensor` a public function so that we can create temporary intermediate tensors in operators that need to do so. (Such as NMS that is implemented above in this stack) - Renames `free_broadcast_tensor` to a more generic name `free_tensor` Differential Revision: D55577026
|
This pull request was exported from Phabricator. Differential Revision: D55577026 |
f70550f to
af92d4e
Compare
Summary:
This diff does a couple of things:
make_tensora public function so that we can create temporary intermediate tensors in operators that need to do so. (Such as NMS that is implemented above in this stack)free_broadcast_tensorto a more generic namefree_tensorDifferential Revision: D55577026