Nicolas De Carli 92284fb2ff Add SVE128 ISA (#158932) ...

Summary: Partly Importing and adapting https://github.com/pytorch/pytorch/pull/138388, adding SVE128 as ISA.

Intention is to add SVE128 translation layers for Vectorized data types.
Idea is to have 1 PR per file, aside from the current one, plus a last one modifying cmake files to enable the new ISA selectively.

Tested current changes on a nightly run, to verify no regressions occur on systems leveraging SVE256.

No regressions spotted when running test_ops.py, a set of 34k unit tests. A machine leveraging SVE128 was used towards this testing.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/158932
Approved by: https://github.com/malfet

2025-09-29 14:49:19 +00:00

..

cuda

[NFC] fixed typo in sparse semi structured filename (#163904)

2025-09-27 21:19:48 +00:00

eigen

[ATen][CPU][Sparse] Use Third-Party Eigen for sparse add and addmm (#155357)

2025-08-23 19:03:55 +00:00

mps

[MPS] mps sparse mul op implementation (#162349)

2025-09-11 18:36:24 +00:00

FlattenIndicesCommon.h

[BE][Sparse] Get rid of gcc-5 workaround (#143653)

2024-12-20 18:40:45 +00:00

FlattenIndicesKernel.cpp

Add SVE128 ISA (#158932)

2025-09-29 14:49:19 +00:00

Macros.h

Faster mul(sparse, sparse) with broadcasting in dense dims. (#85336)

2022-09-23 23:31:19 +00:00

ParamUtils.cpp

torch.sparse.softmax: allow negative dim (#102172)

2023-05-24 19:43:47 +00:00

ParamUtils.h

[BE] Use nested namespaces for sparse (#117415)

2024-01-13 19:51:28 +00:00

SoftMax.cpp

fix torch.sparse.log_softmax on CPU (#161959)

2025-09-11 07:52:05 +00:00

SparseBinaryOpIntersectionCommon.h

[BE][2/5] fix typos in aten/ (aten/src/ATen/native/) (#157551)

2025-07-17 12:08:33 +00:00

SparseBinaryOpIntersectionKernel.cpp

Add SVE128 ISA (#158932)

2025-09-29 14:49:19 +00:00

SparseBlas.cpp

[4/N] Change #include <c10/util/Optional.h> to #include <optional> (#130329)

2024-07-10 01:26:50 +00:00

SparseBlas.h

[BE] Use nested namespaces for sparse (#117415)

2024-01-13 19:51:28 +00:00

SparseBlasImpl.cpp

[ATen][CPU][Sparse] Use Third-Party Eigen for sparse add and addmm (#155357)

2025-08-23 19:03:55 +00:00

SparseBlasImpl.h

[BE] Use nested namespaces for sparse (#117415)

2024-01-13 19:51:28 +00:00

SparseCsrTensor.cpp

Add optional check_pinning argument to _validate_sparse_compressed_tensor/coo_args (#154759)

2025-06-02 10:17:07 +00:00

SparseCsrTensorMath.cpp

[ATen][CPU][Sparse] Use Third-Party Eigen for sparse add and addmm (#155357)

2025-08-23 19:03:55 +00:00

SparseCsrTensorMath.h

c10::string_view -> std::string_view in aten (#141903)

2024-12-07 23:23:52 +00:00

SparseFactories.cpp

[codemod] c10:optional -> std::optional (#126135)

2024-05-14 19:35:51 +00:00

SparseFactories.h

[3/N] Fix Wextra-semi warnings (#139165)

2024-10-30 02:08:13 +00:00

SparseMatMul.cpp

C10_UNUSED to [[maybe_unused]] (#6357) (#138364)

2024-10-19 13:17:43 +00:00

SparseStubs.h

[3/N] Fix Wextra-semi warnings (#139165)

2024-10-30 02:08:13 +00:00

SparseTensor.cpp

Update misleading torch.sparse_coo_tensor error check (#161900)

2025-09-10 19:57:11 +00:00

SparseTensorMath.cpp

[2/N] Use internal linkage in aten C++ files (#151070)

2025-04-14 16:07:17 +00:00

SparseTensorMath.h

[BE] Use nested namespaces for sparse (#117415)

2024-01-13 19:51:28 +00:00

SparseUnaryOps.cpp

[4/N] Fix Wextra-semi warning (#139256)

2024-10-31 03:01:14 +00:00

ValidateCompressedIndicesCommon.h

[BE][2/5] fix typos in aten/ (aten/src/ATen/native/) (#157551)

2025-07-17 12:08:33 +00:00

ValidateCompressedIndicesKernel.cpp

[Reland2] fix missing-prototypes warnings in torch_cpu (Part 5) (#102931)

2023-06-09 21:58:51 +00:00