pytorch/torchgen
Wonjoo Lee a171b0636a Add use_lazy_shape flag to GenLazyIr class (#88444)
Add use_lazy_shape flag to GenLazyIr class to allow XLA to use its custom shape class. The default value is kept to use lazy shape, so this PR does not introduce any new behaviors.

PyTorch/XLA companion PR: https://github.com/pytorch/xla/pull/4111
Pull Request resolved: https://github.com/pytorch/pytorch/pull/88444
Approved by: https://github.com/alanwaketan, https://github.com/wconstab
2022-11-04 08:23:56 +00:00
..
api Reland 2 Many symintifications (#87604) (#87980) 2022-10-28 13:40:11 +00:00
decompositions
dest Add use_lazy_shape flag to GenLazyIr class (#88444) 2022-11-04 08:23:56 +00:00
operator_versions Apply ufmt linter to all py files under torchgen (#81570) 2022-07-16 03:52:25 +00:00
selective_build [torchgen] Fix selective build error on custom namespace (#83141) 2022-08-10 21:27:05 +00:00
shape_functions [JIT SSA] Allow updating shape functions without recompilation (#83629) 2022-08-22 18:03:44 +00:00
static_runtime [static-runtime] run codegen (#87534) 2022-10-25 23:48:16 +00:00
__init__.py Package config/template files with torchgen (#78942) 2022-06-07 13:33:55 +00:00
BUCK.oss rename BUILD.buck to BUCK.oss (#78792) 2022-06-03 07:23:16 +00:00
BUILD.bazel
build.bzl
code_template.py Apply ufmt linter to all py files under torchgen (#81570) 2022-07-16 03:52:25 +00:00
context.py Change API type Tensor[] for structured kernels. (#73350) 2022-09-26 21:46:38 +00:00
gen_backend_stubs.py Revert "Revert "Symintifying slice ops (#85196)"" (#85746) 2022-09-28 04:37:35 +00:00
gen_functionalization_type.py functionalization: make view_copy outputs always contiguous (#85747) 2022-10-21 17:42:02 +00:00
gen_lazy_tensor.py Add use_lazy_shape flag to GenLazyIr class (#88444) 2022-11-04 08:23:56 +00:00
gen_vmap_plumbing.py Move functorch C++ into aten/src/ATen/functorch (#85381) 2022-09-21 15:50:44 +00:00
gen.py [codegen] using TORCH_LIBRARY_FRAGMENT for some namespaces (#88229) 2022-11-03 02:30:02 +00:00
local.py Change API type Tensor[] for structured kernels. (#73350) 2022-09-26 21:46:38 +00:00
model.py [codegen] using TORCH_LIBRARY_FRAGMENT for some namespaces (#88229) 2022-11-03 02:30:02 +00:00
native_function_generation.py [composite compliance] ctc_loss (#84752) 2022-09-22 00:21:11 +00:00
utils.py autograd: fix non-deterministic output in codegen comments (#84695) 2022-09-13 18:41:15 +00:00