Closed
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/160236
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 3a34b08 with merge base af8c232 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
Contributor
Author
|
This is ready for review! |
Contributor
Author
|
Comments have been addressed. I think I'll handle csrc removal next PR. |
Contributor
Author
|
@pytorchbot merge |
Collaborator
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Collaborator
Merge failedReason: 1 jobs have failed, first few of them are: linux-aarch64 / linux-jammy-aarch64-py3.10 / build Details for Dev Infra teamRaised by workflow job |
Contributor
Author
|
@pytorchbot merge |
Collaborator
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
pytorchmergebot
pushed a commit
that referenced
this pull request
Sep 21, 2025
Signed-off-by: Edward Yang <ezyang@meta.com> Pull Request resolved: #163340 Approved by: https://github.com/aorenste ghstack dependencies: #160236
mansiag05
pushed a commit
to mansiag05/pytorch
that referenced
this pull request
Sep 22, 2025
The big semantic change (and the reason for this port) is that we no longer monkeypatch Tensor with torchdim's special methods. The new algorithm for handling dispatch is that we first land in `__torch_function__` and we see if a special FCD implementation needs to be dispatch to first, and if there is nothing we fallback to the standard level strategy. Because there is no longer C binding equivalent of classes, we've condensed _C.Dim and Dim together, and similar for Tensor. This resulted in some bugs as the Python API is sometimes different from the C API. I've attempted to disambiguate these but there may still be mistakes (many early bugs were due to this problem). Dim and DimEntry are especially painful as Dim must abide by Tensor equality semantics, but is pointer equality in C (DimEntry doesn't have this problem). Another difference between C/Python that is subtle is we no longer get implicit conversions from Dim to DimEntry, this also caused some bugs. Much of the mechanical porting work was done by claude code. I have a separate PR that deletes functorch._C, but it was useful having dim.cpp to point claude at it so I haven't done it in this PR. From a reviewing perspective, I need to re-review that I didn't forget to port anything, some noticeably missing "small" things are patched_dim_method. I am still in progress of carefully doing a side-by-side review of ports; "simplifications" from claude code were also a major source of bugs. There are two major feature gaps in the implementation: - DelayedTensor and dot handling are not implemented yet. This should be reasonably easy, just need to do it. However, for the purposes of sharded propagation it is actually better not to reconstruct matmuls. - Splitting dimensions with an index like `[x, y]` doesn't work. The problem is that `__getitem__` interprets this as advanced indexing and sends the list to torch.tensor to turn into a tensor, instead of being eligible for `__torch_function__`. I think I might need to hard code a special case for this or something? Signed-off-by: Edward Yang <ezyang@meta.com> Pull Request resolved: pytorch#160236 Approved by: https://github.com/zdevito, https://github.com/albanD
mansiag05
pushed a commit
to mansiag05/pytorch
that referenced
this pull request
Sep 22, 2025
Signed-off-by: Edward Yang <ezyang@meta.com> Pull Request resolved: pytorch#163340 Approved by: https://github.com/aorenste ghstack dependencies: pytorch#160236
cleonard530
pushed a commit
to cleonard530/pytorch
that referenced
this pull request
Sep 22, 2025
The big semantic change (and the reason for this port) is that we no longer monkeypatch Tensor with torchdim's special methods. The new algorithm for handling dispatch is that we first land in `__torch_function__` and we see if a special FCD implementation needs to be dispatch to first, and if there is nothing we fallback to the standard level strategy. Because there is no longer C binding equivalent of classes, we've condensed _C.Dim and Dim together, and similar for Tensor. This resulted in some bugs as the Python API is sometimes different from the C API. I've attempted to disambiguate these but there may still be mistakes (many early bugs were due to this problem). Dim and DimEntry are especially painful as Dim must abide by Tensor equality semantics, but is pointer equality in C (DimEntry doesn't have this problem). Another difference between C/Python that is subtle is we no longer get implicit conversions from Dim to DimEntry, this also caused some bugs. Much of the mechanical porting work was done by claude code. I have a separate PR that deletes functorch._C, but it was useful having dim.cpp to point claude at it so I haven't done it in this PR. From a reviewing perspective, I need to re-review that I didn't forget to port anything, some noticeably missing "small" things are patched_dim_method. I am still in progress of carefully doing a side-by-side review of ports; "simplifications" from claude code were also a major source of bugs. There are two major feature gaps in the implementation: - DelayedTensor and dot handling are not implemented yet. This should be reasonably easy, just need to do it. However, for the purposes of sharded propagation it is actually better not to reconstruct matmuls. - Splitting dimensions with an index like `[x, y]` doesn't work. The problem is that `__getitem__` interprets this as advanced indexing and sends the list to torch.tensor to turn into a tensor, instead of being eligible for `__torch_function__`. I think I might need to hard code a special case for this or something? Signed-off-by: Edward Yang <ezyang@meta.com> Pull Request resolved: pytorch#160236 Approved by: https://github.com/zdevito, https://github.com/albanD
cleonard530
pushed a commit
to cleonard530/pytorch
that referenced
this pull request
Sep 22, 2025
Signed-off-by: Edward Yang <ezyang@meta.com> Pull Request resolved: pytorch#163340 Approved by: https://github.com/aorenste ghstack dependencies: pytorch#160236
dsashidh
pushed a commit
to dsashidh/pytorch
that referenced
this pull request
Sep 26, 2025
The big semantic change (and the reason for this port) is that we no longer monkeypatch Tensor with torchdim's special methods. The new algorithm for handling dispatch is that we first land in `__torch_function__` and we see if a special FCD implementation needs to be dispatch to first, and if there is nothing we fallback to the standard level strategy. Because there is no longer C binding equivalent of classes, we've condensed _C.Dim and Dim together, and similar for Tensor. This resulted in some bugs as the Python API is sometimes different from the C API. I've attempted to disambiguate these but there may still be mistakes (many early bugs were due to this problem). Dim and DimEntry are especially painful as Dim must abide by Tensor equality semantics, but is pointer equality in C (DimEntry doesn't have this problem). Another difference between C/Python that is subtle is we no longer get implicit conversions from Dim to DimEntry, this also caused some bugs. Much of the mechanical porting work was done by claude code. I have a separate PR that deletes functorch._C, but it was useful having dim.cpp to point claude at it so I haven't done it in this PR. From a reviewing perspective, I need to re-review that I didn't forget to port anything, some noticeably missing "small" things are patched_dim_method. I am still in progress of carefully doing a side-by-side review of ports; "simplifications" from claude code were also a major source of bugs. There are two major feature gaps in the implementation: - DelayedTensor and dot handling are not implemented yet. This should be reasonably easy, just need to do it. However, for the purposes of sharded propagation it is actually better not to reconstruct matmuls. - Splitting dimensions with an index like `[x, y]` doesn't work. The problem is that `__getitem__` interprets this as advanced indexing and sends the list to torch.tensor to turn into a tensor, instead of being eligible for `__torch_function__`. I think I might need to hard code a special case for this or something? Signed-off-by: Edward Yang <ezyang@meta.com> Pull Request resolved: pytorch#160236 Approved by: https://github.com/zdevito, https://github.com/albanD
dsashidh
pushed a commit
to dsashidh/pytorch
that referenced
this pull request
Sep 26, 2025
Signed-off-by: Edward Yang <ezyang@meta.com> Pull Request resolved: pytorch#163340 Approved by: https://github.com/aorenste ghstack dependencies: pytorch#160236
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
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):
The big semantic change (and the reason for this port) is that we no longer monkeypatch Tensor with torchdim's special methods. The new algorithm for handling dispatch is that we first land in
__torch_function__and we see if a special FCD implementation needs to be dispatch to first, and if there is nothing we fallback to the standard level strategy.Because there is no longer C binding equivalent of classes, we've condensed _C.Dim and Dim together, and similar for Tensor. This resulted in some bugs as the Python API is sometimes different from the C API. I've attempted to disambiguate these but there may still be mistakes (many early bugs were due to this problem). Dim and DimEntry are especially painful as Dim must abide by Tensor equality semantics, but is pointer equality in C (DimEntry doesn't have this problem). Another difference between C/Python that is subtle is we no longer get implicit conversions from Dim to DimEntry, this also caused some bugs.
Much of the mechanical porting work was done by claude code. I have a separate PR that deletes functorch._C, but it was useful having dim.cpp to point claude at it so I haven't done it in this PR. From a reviewing perspective, I need to re-review that I didn't forget to port anything, some noticeably missing "small" things are patched_dim_method. I am still in progress of carefully doing a side-by-side review of ports; "simplifications" from claude code were also a major source of bugs.
There are two major feature gaps in the implementation:
[x, y]doesn't work. The problem is that__getitem__interprets this as advanced indexing and sends the list to torch.tensor to turn into a tensor, instead of being eligible for__torch_function__. I think I might need to hard code a special case for this or something?Signed-off-by: Edward Yang ezyang@meta.com
cc @albanD