torch-mlir/test/python/importer/jit_ir
Sean Silva 140babd952 Add minimal support for Union types.
A recent PyTorch commit made ConstantPad2d call a helper function with a
`Union[int, float]` type annotated. This commit adds minimal support for
representing and dealing with that.
https://github.com/pytorch/pytorch/pull/73287

Changes:
- Adding support for `!torch.union<T1, T2, T3>`/`Torch::UnionType`,
  along with the importer and CAPI code.
- Add support in isValidSubtype for union types.
- Adding a canonicalizer for `torch.derefine` to help simplify some code
  that derefines to a UnionType (this also fixes #664).

There is still more work to do for really supporting UnionType well,
such as canonicalizing UnionType's so that they can be compared with
pointer equality.
2022-03-29 17:45:48 -07:00
..
ivalue_import Elide `!torch.` prefix in nested dialect types. 2022-03-15 17:24:08 -07:00
node_import Add minimal support for Union types. 2022-03-29 17:45:48 -07:00
get_registered_ops.py Move external/torch-mlir to the root of the repo. 2021-09-27 17:11:08 -07:00
lit.local.cfg Move external/torch-mlir to the root of the repo. 2021-09-27 17:11:08 -07:00