Natalia Gimelshein 37c6087334 Add split-K control to cuBLAS reduced-precision settings (#164766) ... ## Summary - add a CuBLASReductionOption enum so the CUDA context can track reduced-precision and split-K options - extend the Python bindings, backend helpers, and docs to accept an optional allow_splitk argument for fp16/bf16 matmul controls - update cuBLAS/cuBLASLt call sites plus dynamo guards and tests to respect the new combinations ## Testing - python test/test_cuda.py TestCuda.test_cublas_allow_fp16_reduced_precision_reduction_get_set -v *(fails: ModuleNotFoundError: No module named 'psutil')* ------ https://chatgpt.com/codex/tasks/task_e_68e404623178832f8a3e1d34e1e175da Pull Request resolved: https://github.com/pytorch/pytorch/pull/164766 Approved by: https://github.com/malfet, https://github.com/albanD		2025-10-08 18:48:45 +00:00
..
_coreml
_nnapi	Add pyrefly suppressions 2/n (#164513)	2025-10-03 02:46:13 +00:00
cpu	Add SVE128 ISA (#158932)	2025-09-29 14:49:19 +00:00
cuda	Add split-K control to cuBLAS reduced-precision settings (#164766)	2025-10-08 18:48:45 +00:00
cudnn	Add pyrefly suppressions 2/n (#164513)	2025-10-03 02:46:13 +00:00
cusparselt	Add pyrefly suppressions 2/n (#164513)	2025-10-03 02:46:13 +00:00
kleidiai
mha
miopen	[ROCm] add flag torch.backends.miopen.immediate (#158951)	2025-07-25 04:01:51 +00:00
mkl
mkldnn
mps	[MPS] Add API to query GPU core count (#160414)	2025-08-14 00:05:17 +00:00
nnpack
openmp
opt_einsum	[2/N] Fix ruff warnings (#164460)	2025-10-04 03:40:32 +00:00
quantized
xeon	Add pyrefly suppressions 2/n (#164513)	2025-10-03 02:46:13 +00:00
xnnpack
__init__.py	[ROCm] add flag torch.backends.miopen.immediate (#158951)	2025-07-25 04:01:51 +00:00