FFFrog 68cb854d73 Fix CPUFallback Mechinasm on TensorList Type (#105209) ...

Fixes #104965

Currently, the cpufallback mechinasm lack the code logic of TensorList, so some operators like _foreach_add_/_foreach_add don`t work well.

cc  @bdhirsh

Pull Request resolved: https://github.com/pytorch/pytorch/pull/105209
Approved by: https://github.com/bdhirsh

2023-08-05 15:38:30 +00:00

..

no_python_abi_suffix_test

…

self_compiler_include_dirs_test

…

torch_test_cpp_extension

…

cpp_c10d_extension.cpp

…

cpp_c10d_extension.hpp

…

cpp_frontend_extension.cpp

…

cublas_extension.cpp

…

cuda_dlink_extension_add.cu

…

cuda_dlink_extension_add.cuh

…

cuda_dlink_extension_kernel.cu

…

cuda_dlink_extension.cpp

…

cuda_extension_kernel2.cu

…

cuda_extension_kernel.cu

…

cuda_extension.cpp

…

cuda_extension.cu

…

cudnn_extension.cpp

…

cusolver_extension.cpp

…

dangling_impl_extension.cpp

…

doubler.h

…

extension.cpp

…

identity.cpp

[functorch] disable C++ Function under functorch transforms (#103957)

2023-06-23 11:01:44 +00:00

jit_extension2.cpp

…

jit_extension.cpp

…

mps_extension.mm

…

open_registration_extension.cpp

Fix CPUFallback Mechinasm on TensorList Type (#105209)

2023-08-05 15:38:30 +00:00

ort_extension.cpp

…

rng_extension.cpp

…

setup.py

…

torch_library.cu

…