aoti_torch__reinterpret_tensor #14688
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Summary: This diff introduce `aoti_torch_delete_tensor_object` for deleting tensors created during cuda backend inference. Reviewed By: [ghstack-poisoned]
Summary: This function introduce aoti_torch_create_tensor_from_blob_v2, a function that create tensor from data blob and custom stride and size. Worth to notice that unlike aoti_torch_empty_strided, the tensor created by aoti_torch_create_tensor_from_blob_v2 will not have the control of the memory blob. Therefore when we delete it, the memory will not be freed. Reviewed By: Differential Revision: [ghstack-poisoned]
Summary: Introduced aoti_torch__reinterpret_tensor, which creates a new tensor view that reinterprets the same underlying memory with custom shape and strides. Reviewed By: Differential Revision: [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/14688
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New Failure, 27 Pending, 1 Unrelated FailureAs of commit b4adeda with merge base 44972ad ( FLAKY - The following job failed but was likely due to flakiness present on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This was referenced Sep 30, 2025
Merged
Merged
meta-cla
bot
added
the
CLA Signed
manuelcandales
approved these changes
Sep 30, 2025
larryliu0820
added a commit
that referenced
this pull request
Sep 30, 2025
Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * #14690 * #14689 * #14688 * __->__ #14687 * #14686 Summary: This function introduce aoti_torch_create_tensor_from_blob_v2, a function that create tensor from data blob and custom stride and size. Worth to notice that unlike aoti_torch_empty_strided, the tensor created by aoti_torch_create_tensor_from_blob_v2 will not have the control of the memory blob. Therefore when we delete it, the memory will not be freed. Reviewed By: Differential Revision:
larryliu0820
added a commit
that referenced
this pull request
Sep 30, 2025
Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * #14690 * #14689 * #14688 * __->__ #14700 * #14686 Summary: This is a manual cherry pick of #14687 This function introduce aoti_torch_create_tensor_from_blob_v2, a function that create tensor from data blob and custom stride and size. Worth to notice that unlike aoti_torch_empty_strided, the tensor created by aoti_torch_create_tensor_from_blob_v2 will not have the control of the memory blob. Therefore when we delete it, the memory will not be freed. Reviewed By: Differential Revision:
larryliu0820
added a commit
that referenced
this pull request
Sep 30, 2025
Stack from [ghstack](https://github.com/ezyang/ghstack) (oldest at bottom): * #14690 * __->__ #14689 * #14688 * #14687 * #14686 Summary: This diff introduce `aoti_torch_copy_`, the function for copying tensor inside cuda backend. Right now it only support copy between tensors with same dtype. Reviewed By: Differential Revision:
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Stack from ghstack (oldest at bottom):
Summary:
Introduced aoti_torch__reinterpret_tensor, which creates a new tensor view that reinterprets the same underlying memory with custom shape and strides.
Reviewed By:
Differential Revision: