[slimtensor] Add SlimTensor class with basic properties and CPU copy operation #16385

Gasoonjia · 2025-12-24T01:07:55Z

Stack from ghstack (oldest at bottom):

-> [slimtensor] Add SlimTensor class with basic properties and CPU copy operation #16385
[slimtensor] Add SizesAndStrides and dimension utility functions for slimtensor usage #16384
[slimtensor] Storage and SharedPtr for CPU owning mode #16383
[slimtensor] Introduce Device and ScalarType headers for SlimTensor minimal support #16382

Key components:

c10/core/Contiguity.h - Contiguity checking utility:
- _compute_contiguous<T>() - computes whether a tensor with given sizes/strides is contiguous in memory (row-major order)
core/SlimTensor.h - Main SlimTensor class with:
- Constructors: Default (undefined tensor) and full constructor with storage, sizes, strides, dtype, and storage_offset
- Property accessors:
  - sizes(), size(dim) - get tensor dimensions with negative indexing support
  - strides(), stride(dim) - get tensor strides with negative indexing support
  - dtype(), device(), device_type(), device_index()
  - numel(), dim(), nbytes(), itemsize()
  - data_ptr() - returns pointer to tensor data (adjusted for storage_offset)
  - storage_offset(), storage()
- State queries: defined(), is_cpu(), is_contiguous(), is_empty()
- Copy operation: copy_(other) - copies data from another tensor
  - Fast path: uses memcpy for both-contiguous tensors
  - Slow path: element-wise copy respecting strides for non-contiguous tensors
- Setters: reset(), set_storage(), set_sizes_and_strides()

Curretnt constraints:

Only CPU device supported
Only Float32 dtype tested
copy_() only supports CPU-to-CPU copy

Those contraints will be further improved in the following diffs

Differential Revision: D89750150

…operation **Key components:** 1. **`c10/core/Contiguity.h`** - Contiguity checking utility: - `_compute_contiguous<T>()` - computes whether a tensor with given sizes/strides is contiguous in memory (row-major order) 2. **`core/SlimTensor.h`** - Main SlimTensor class with: - **Constructors**: Default (undefined tensor) and full constructor with storage, sizes, strides, dtype, and storage_offset - **Property accessors**: - `sizes()`, `size(dim)` - get tensor dimensions with negative indexing support - `strides()`, `stride(dim)` - get tensor strides with negative indexing support - `dtype()`, `device()`, `device_type()`, `device_index()` - `numel()`, `dim()`, `nbytes()`, `itemsize()` - `data_ptr()` - returns pointer to tensor data (adjusted for storage_offset) - `storage_offset()`, `storage()` - **State queries**: `defined()`, `is_cpu()`, `is_contiguous()`, `is_empty()` - **Copy operation**: `copy_(other)` - copies data from another tensor - Fast path: uses memcpy for both-contiguous tensors - Slow path: element-wise copy respecting strides for non-contiguous tensors - **Setters**: `reset()`, `set_storage()`, `set_sizes_and_strides()` **Curretnt constraints:** - Only CPU device supported - Only Float32 dtype tested - copy_() only supports CPU-to-CPU copy Those contraints will be further improved in the following diffs Differential Revision: [D89750150](https://our.internmc.facebook.com/intern/diff/D89750150/) [ghstack-poisoned]

pytorch-bot · 2025-12-24T01:07:59Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/16385

📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

❌ 2 New Failures, 5 Unrelated Failures

As of commit 55d69cd with merge base 9ba1b5d ():

NEW FAILURES - The following jobs have failed:

pull / unittest-editable / macos / macos-job (gh)
export/tests/test_target_recipes.py::TestTargetRecipes::test_mv3_model
Test CUDA Builds / test-model-cuda-e2e (mistralai, Voxtral-Mini-3B-2507, quantized-int4-tile-packed) / linux-job (gh)
RuntimeError: Command docker exec -t e6cfb8a93b9c8bd25a56c2f8182fb6f0e966c490692633948d0a06bf494a8e63 /exec failed with exit code 1

BROKEN TRUNK - The following jobs failed but were present on the merge base:

👉 Rebase onto the `viable/strict` branch to avoid these failures

Test CUDA Windows Export / export-model-cuda-windows-artifact (openai, whisper-large-v3-turbo, non-quantized) / linux-job (gh) (trunk failure)
/opt/conda/lib/python3.10/site-packages/torch/include/ATen/core/PhiloxRNGEngine.h:135:79: error: ‘M_PI’ was not declared in this scope
Test CUDA Windows Export / export-model-cuda-windows-artifact (openai, whisper-large-v3-turbo, quantized-int4-weight-only) / linux-job (gh) (trunk failure)
/opt/conda/lib/python3.10/site-packages/torch/include/ATen/core/PhiloxRNGEngine.h:135:79: error: ‘M_PI’ was not declared in this scope
Test CUDA Windows Export / export-model-cuda-windows-artifact (openai, whisper-small, non-quantized) / linux-job (gh) (trunk failure)
/opt/conda/lib/python3.10/site-packages/torch/include/ATen/core/PhiloxRNGEngine.h:135:79: error: ‘M_PI’ was not declared in this scope
Test CUDA Windows Export / export-model-cuda-windows-artifact (openai, whisper-small, quantized-int4-weight-only) / linux-job (gh) (trunk failure)
/opt/conda/lib/python3.10/site-packages/torch/include/ATen/core/PhiloxRNGEngine.h:135:79: error: ‘M_PI’ was not declared in this scope

UNSTABLE - The following job is marked as unstable, possibly due to flakiness on trunk:

pull / android / run-emulator (gh) (#16137)
Timeout waiting for emulator to boot.

This comment was automatically generated by Dr. CI and updates every 15 minutes.

…operation **Key components:** 1. **`c10/core/Contiguity.h`** - Contiguity checking utility: - `_compute_contiguous<T>()` - computes whether a tensor with given sizes/strides is contiguous in memory (row-major order) 2. **`core/SlimTensor.h`** - Main SlimTensor class with: - **Constructors**: Default (undefined tensor) and full constructor with storage, sizes, strides, dtype, and storage_offset - **Property accessors**: - `sizes()`, `size(dim)` - get tensor dimensions with negative indexing support - `strides()`, `stride(dim)` - get tensor strides with negative indexing support - `dtype()`, `device()`, `device_type()`, `device_index()` - `numel()`, `dim()`, `nbytes()`, `itemsize()` - `data_ptr()` - returns pointer to tensor data (adjusted for storage_offset) - `storage_offset()`, `storage()` - **State queries**: `defined()`, `is_cpu()`, `is_contiguous()`, `is_empty()` - **Copy operation**: `copy_(other)` - copies data from another tensor - Fast path: uses memcpy for both-contiguous tensors - Slow path: element-wise copy respecting strides for non-contiguous tensors - **Setters**: `reset()`, `set_storage()`, `set_sizes_and_strides()` **Curretnt constraints:** - Only CPU device supported - Only Float32 dtype tested - copy_() only supports CPU-to-CPU copy Those contraints will be further improved in the following diffs Differential Revision: [D89750150](https://our.internmc.facebook.com/intern/diff/D89750150/) ghstack-source-id: 331051343 Pull Request resolved: #16385

github-actions · 2025-12-24T01:09:07Z

This PR needs a `release notes:` label

If your change should be included in the release notes (i.e. would users of this library care about this change?), please use a label starting with release notes:. This helps us keep track and include your important work in the next release notes.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "release notes: none"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

…d CPU copy operation" **Key components:** 1. **`c10/core/Contiguity.h`** - Contiguity checking utility: - `_compute_contiguous<T>()` - computes whether a tensor with given sizes/strides is contiguous in memory (row-major order) 2. **`core/SlimTensor.h`** - Main SlimTensor class with: - **Constructors**: Default (undefined tensor) and full constructor with storage, sizes, strides, dtype, and storage_offset - **Property accessors**: - `sizes()`, `size(dim)` - get tensor dimensions with negative indexing support - `strides()`, `stride(dim)` - get tensor strides with negative indexing support - `dtype()`, `device()`, `device_type()`, `device_index()` - `numel()`, `dim()`, `nbytes()`, `itemsize()` - `data_ptr()` - returns pointer to tensor data (adjusted for storage_offset) - `storage_offset()`, `storage()` - **State queries**: `defined()`, `is_cpu()`, `is_contiguous()`, `is_empty()` - **Copy operation**: `copy_(other)` - copies data from another tensor - Fast path: uses memcpy for both-contiguous tensors - Slow path: element-wise copy respecting strides for non-contiguous tensors - **Setters**: `reset()`, `set_storage()`, `set_sizes_and_strides()` **Curretnt constraints:** - Only CPU device supported - Only Float32 dtype tested - copy_() only supports CPU-to-CPU copy Those contraints will be further improved in the following diffs Differential Revision: [D89750150](https://our.internmc.facebook.com/intern/diff/D89750150/) [ghstack-poisoned]

…operation Pull Request resolved: #16385 **Key components:** 1. **`c10/core/Contiguity.h`** - Contiguity checking utility: - `_compute_contiguous<T>()` - computes whether a tensor with given sizes/strides is contiguous in memory (row-major order) 2. **`core/SlimTensor.h`** - Main SlimTensor class with: - **Constructors**: Default (undefined tensor) and full constructor with storage, sizes, strides, dtype, and storage_offset - **Property accessors**: - `sizes()`, `size(dim)` - get tensor dimensions with negative indexing support - `strides()`, `stride(dim)` - get tensor strides with negative indexing support - `dtype()`, `device()`, `device_type()`, `device_index()` - `numel()`, `dim()`, `nbytes()`, `itemsize()` - `data_ptr()` - returns pointer to tensor data (adjusted for storage_offset) - `storage_offset()`, `storage()` - **State queries**: `defined()`, `is_cpu()`, `is_contiguous()`, `is_empty()` - **Copy operation**: `copy_(other)` - copies data from another tensor - Fast path: uses memcpy for both-contiguous tensors - Slow path: element-wise copy respecting strides for non-contiguous tensors - **Setters**: `reset()`, `set_storage()`, `set_sizes_and_strides()` **Curretnt constraints:** - Only CPU device supported - Only Float32 dtype tested - copy_() only supports CPU-to-CPU copy Those contraints will be further improved in the following diffs ghstack-source-id: 331187116 @exported-using-ghexport Differential Revision: [D89750150](https://our.internmc.facebook.com/intern/diff/D89750150/)

…d CPU copy operation" **Key components:** 1. **`c10/core/Contiguity.h`** - Contiguity checking utility: - `_compute_contiguous<T>()` - computes whether a tensor with given sizes/strides is contiguous in memory (row-major order) 2. **`core/SlimTensor.h`** - Main SlimTensor class with: - **Constructors**: Default (undefined tensor) and full constructor with storage, sizes, strides, dtype, and storage_offset - **Property accessors**: - `sizes()`, `size(dim)` - get tensor dimensions with negative indexing support - `strides()`, `stride(dim)` - get tensor strides with negative indexing support - `dtype()`, `device()`, `device_type()`, `device_index()` - `numel()`, `dim()`, `nbytes()`, `itemsize()` - `data_ptr()` - returns pointer to tensor data (adjusted for storage_offset) - `storage_offset()`, `storage()` - **State queries**: `defined()`, `is_cpu()`, `is_contiguous()`, `is_empty()` - **Copy operation**: `copy_(other)` - copies data from another tensor - Fast path: uses memcpy for both-contiguous tensors - Slow path: element-wise copy respecting strides for non-contiguous tensors - **Setters**: `reset()`, `set_storage()`, `set_sizes_and_strides()` **Curretnt constraints:** - Only CPU device supported - Only Float32 dtype tested - copy_() only supports CPU-to-CPU copy Those contraints will be further improved in the following diffs Differential Revision: [D89750150](https://our.internmc.facebook.com/intern/diff/D89750150/) [ghstack-poisoned]

…operation Pull Request resolved: #16385 **Key components:** 1. **`c10/core/Contiguity.h`** - Contiguity checking utility: - `_compute_contiguous<T>()` - computes whether a tensor with given sizes/strides is contiguous in memory (row-major order) 2. **`core/SlimTensor.h`** - Main SlimTensor class with: - **Constructors**: Default (undefined tensor) and full constructor with storage, sizes, strides, dtype, and storage_offset - **Property accessors**: - `sizes()`, `size(dim)` - get tensor dimensions with negative indexing support - `strides()`, `stride(dim)` - get tensor strides with negative indexing support - `dtype()`, `device()`, `device_type()`, `device_index()` - `numel()`, `dim()`, `nbytes()`, `itemsize()` - `data_ptr()` - returns pointer to tensor data (adjusted for storage_offset) - `storage_offset()`, `storage()` - **State queries**: `defined()`, `is_cpu()`, `is_contiguous()`, `is_empty()` - **Copy operation**: `copy_(other)` - copies data from another tensor - Fast path: uses memcpy for both-contiguous tensors - Slow path: element-wise copy respecting strides for non-contiguous tensors - **Setters**: `reset()`, `set_storage()`, `set_sizes_and_strides()` **Curretnt constraints:** - Only CPU device supported - Only Float32 dtype tested - copy_() only supports CPU-to-CPU copy Those contraints will be further improved in the following diffs ghstack-source-id: 332731275 @exported-using-ghexport Differential Revision: [D89750150](https://our.internmc.facebook.com/intern/diff/D89750150/)

…d CPU copy operation" **Key components:** 1. **`c10/core/Contiguity.h`** - Contiguity checking utility: - `_compute_contiguous<T>()` - computes whether a tensor with given sizes/strides is contiguous in memory (row-major order) 2. **`core/SlimTensor.h`** - Main SlimTensor class with: - **Constructors**: Default (undefined tensor) and full constructor with storage, sizes, strides, dtype, and storage_offset - **Property accessors**: - `sizes()`, `size(dim)` - get tensor dimensions with negative indexing support - `strides()`, `stride(dim)` - get tensor strides with negative indexing support - `dtype()`, `device()`, `device_type()`, `device_index()` - `numel()`, `dim()`, `nbytes()`, `itemsize()` - `data_ptr()` - returns pointer to tensor data (adjusted for storage_offset) - `storage_offset()`, `storage()` - **State queries**: `defined()`, `is_cpu()`, `is_contiguous()`, `is_empty()` - **Copy operation**: `copy_(other)` - copies data from another tensor - Fast path: uses memcpy for both-contiguous tensors - Slow path: element-wise copy respecting strides for non-contiguous tensors - **Setters**: `reset()`, `set_storage()`, `set_sizes_and_strides()` **Curretnt constraints:** - Only CPU device supported - Only Float32 dtype tested - copy_() only supports CPU-to-CPU copy Those contraints will be further improved in the following diffs Differential Revision: [D89750150](https://our.internmc.facebook.com/intern/diff/D89750150/) [ghstack-poisoned]

…operation Pull Request resolved: #16385 **Key components:** 1. **`c10/core/Contiguity.h`** - Contiguity checking utility: - `_compute_contiguous<T>()` - computes whether a tensor with given sizes/strides is contiguous in memory (row-major order) 2. **`core/SlimTensor.h`** - Main SlimTensor class with: - **Constructors**: Default (undefined tensor) and full constructor with storage, sizes, strides, dtype, and storage_offset - **Property accessors**: - `sizes()`, `size(dim)` - get tensor dimensions with negative indexing support - `strides()`, `stride(dim)` - get tensor strides with negative indexing support - `dtype()`, `device()`, `device_type()`, `device_index()` - `numel()`, `dim()`, `nbytes()`, `itemsize()` - `data_ptr()` - returns pointer to tensor data (adjusted for storage_offset) - `storage_offset()`, `storage()` - **State queries**: `defined()`, `is_cpu()`, `is_contiguous()`, `is_empty()` - **Copy operation**: `copy_(other)` - copies data from another tensor - Fast path: uses memcpy for both-contiguous tensors - Slow path: element-wise copy respecting strides for non-contiguous tensors - **Setters**: `reset()`, `set_storage()`, `set_sizes_and_strides()` **Curretnt constraints:** - Only CPU device supported - Only Float32 dtype tested - copy_() only supports CPU-to-CPU copy Those contraints will be further improved in the following diffs ghstack-source-id: 332735712 @exported-using-ghexport Differential Revision: [D89750150](https://our.internmc.facebook.com/intern/diff/D89750150/)

…d CPU copy operation" **Key components:** 1. **`c10/core/Contiguity.h`** - Contiguity checking utility: - `_compute_contiguous<T>()` - computes whether a tensor with given sizes/strides is contiguous in memory (row-major order) 2. **`core/SlimTensor.h`** - Main SlimTensor class with: - **Constructors**: Default (undefined tensor) and full constructor with storage, sizes, strides, dtype, and storage_offset - **Property accessors**: - `sizes()`, `size(dim)` - get tensor dimensions with negative indexing support - `strides()`, `stride(dim)` - get tensor strides with negative indexing support - `dtype()`, `device()`, `device_type()`, `device_index()` - `numel()`, `dim()`, `nbytes()`, `itemsize()` - `data_ptr()` - returns pointer to tensor data (adjusted for storage_offset) - `storage_offset()`, `storage()` - **State queries**: `defined()`, `is_cpu()`, `is_contiguous()`, `is_empty()` - **Copy operation**: `copy_(other)` - copies data from another tensor - Fast path: uses memcpy for both-contiguous tensors - Slow path: element-wise copy respecting strides for non-contiguous tensors - **Setters**: `reset()`, `set_storage()`, `set_sizes_and_strides()` **Curretnt constraints:** - Only CPU device supported - Only Float32 dtype tested - copy_() only supports CPU-to-CPU copy Those contraints will be further improved in the following diffs Differential Revision: [D89750150](https://our.internmc.facebook.com/intern/diff/D89750150/) [ghstack-poisoned]

…operation Pull Request resolved: #16385 **Key components:** 1. **`c10/core/Contiguity.h`** - Contiguity checking utility: - `_compute_contiguous<T>()` - computes whether a tensor with given sizes/strides is contiguous in memory (row-major order) 2. **`core/SlimTensor.h`** - Main SlimTensor class with: - **Constructors**: Default (undefined tensor) and full constructor with storage, sizes, strides, dtype, and storage_offset - **Property accessors**: - `sizes()`, `size(dim)` - get tensor dimensions with negative indexing support - `strides()`, `stride(dim)` - get tensor strides with negative indexing support - `dtype()`, `device()`, `device_type()`, `device_index()` - `numel()`, `dim()`, `nbytes()`, `itemsize()` - `data_ptr()` - returns pointer to tensor data (adjusted for storage_offset) - `storage_offset()`, `storage()` - **State queries**: `defined()`, `is_cpu()`, `is_contiguous()`, `is_empty()` - **Copy operation**: `copy_(other)` - copies data from another tensor - Fast path: uses memcpy for both-contiguous tensors - Slow path: element-wise copy respecting strides for non-contiguous tensors - **Setters**: `reset()`, `set_storage()`, `set_sizes_and_strides()` **Curretnt constraints:** - Only CPU device supported - Only Float32 dtype tested - copy_() only supports CPU-to-CPU copy Those contraints will be further improved in the following diffs ghstack-source-id: 332983722 @exported-using-ghexport Differential Revision: [D89750150](https://our.internmc.facebook.com/intern/diff/D89750150/)

@Gasoonjia

…operation (#16550) This PR was created by the merge bot to help merge the original PR into the main branch. ghstack PR number: #16385 by @Gasoonjia ^ Please use this as the source of truth for the PR details, comments, and reviews ghstack PR base: https://github.com/pytorch/executorch/tree/gh/gasoonjia/76/base ghstack PR head: https://github.com/pytorch/executorch/tree/gh/gasoonjia/76/head Merge bot PR base: https://github.com/pytorch/executorch/tree/gh/gasoonjia/75/orig Merge bot PR head: https://github.com/pytorch/executorch/tree/gh/gasoonjia/76/orig Differential Revision: [D89750150](https://our.internmc.facebook.com/intern/diff/D89750150/) @diff-train-skip-merge --------- Co-authored-by: gasoonjia <[email protected]> Co-authored-by: Gasoonjia <[email protected]>

This was referenced Dec 24, 2025

[slimtensor] Introduce Device and ScalarType headers for SlimTensor minimal support #16382

Merged

[slimtensor] Storage and SharedPtr for CPU owning mode #16383

Merged

Gasoonjia mentioned this pull request Dec 24, 2025

[slimtensor] Add SizesAndStrides and dimension utility functions for slimtensor usage #16384

Merged

meta-cla bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Dec 24, 2025

meta-codesync bot added fb-exported meta-exported labels Dec 24, 2025

This was referenced Dec 26, 2025

[slimtensor] Add factory functions for creating empty CPU tensors #16398

Open

[slimtensor] Add all required dtype support (Int8/16/32/64, Bool, BFloat16) #16399

Open

JacobSzwejbka approved these changes Jan 9, 2026

View reviewed changes

manuelcandales approved these changes Jan 9, 2026

View reviewed changes

meta-codesync bot merged commit 6b137ae into gh/gasoonjia/76/base Jan 12, 2026
166 of 175 checks passed

meta-codesync bot deleted the gh/gasoonjia/76/head branch January 12, 2026 22:23

meta-codesync bot temporarily deployed to cherry-pick-bot January 12, 2026 22:23 Inactive

pytorchbot mentioned this pull request Jan 12, 2026

[slimtensor] Add SlimTensor class with basic properties and CPU copy operation #16550

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

[slimtensor] Add SlimTensor class with basic properties and CPU copy operation #16385

[slimtensor] Add SlimTensor class with basic properties and CPU copy operation #16385

Uh oh!

Gasoonjia commented Dec 24, 2025 •

edited

Loading

Uh oh!

pytorch-bot bot commented Dec 24, 2025 •

edited

Loading

Uh oh!

github-actions bot commented Dec 24, 2025

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

[slimtensor] Add SlimTensor class with basic properties and CPU copy operation #16385

[slimtensor] Add SlimTensor class with basic properties and CPU copy operation #16385

Uh oh!

Conversation

Gasoonjia commented Dec 24, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

pytorch-bot bot commented Dec 24, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/16385

❌ 2 New Failures, 5 Unrelated Failures

Uh oh!

github-actions bot commented Dec 24, 2025

This PR needs a release notes: label

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Gasoonjia commented Dec 24, 2025 •

edited

Loading

pytorch-bot bot commented Dec 24, 2025 •

edited

Loading

This PR needs a `release notes:` label