Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add link to work log issue in FIX_LOWERING_FOR_CORE_ATEN_OPS.md #5944

Merged
merged 1 commit into from
Nov 29, 2023
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions FIX_LOWERING_FOR_CORE_ATEN_OPS.md
Original file line number Diff line number Diff line change
@@ -1,5 +1,7 @@
In order for PyTorch/XLA to support the PyTorch core ATen opset, it requires lowering each core ATen op in PyTorch/XLA. Note that this document will serve as a guide for fixing these lowering for core aten opset, specifically looking at [test_core_aten_ops.py test](https://github.com/pytorch/xla/blob/master/test/test_core_aten_ops.py). This guide will **not** cover how to lower an op in PyTorch/XLA, please refer our [op lowering guide](https://github.com/pytorch/xla/blob/master/OP_LOWERING_GUIDE.md) for this.

We also have a worklog for lowering these core aten ops in a GitHub issue, so we can track who's working on which ops and share some findings: [[Core Aten ops] Logs for fixing core aten ops coverage issues](https://github.com/pytorch/xla/issues/5934).

Let's go back and take a closer look at the [test_core_aten_ops.py test](https://github.com/pytorch/xla/blob/master/test/test_core_aten_ops.py), which is the source of truth to verify and correctness of these lowerings. The core of this test file is the `run_export_and_compare` at https://github.com/pytorch/xla/blob/master/test/test_core_aten_ops.py#L28. Each op unit test initializes the input and passes the op as a function and its inputs. The `run_export_and_compare` has multiple subtests that have the following structure:
- `torch_eval`
- `torch_xla_eval`
Expand Down
Loading