uv/crates/uv-torch/src
Yu, Guangye b1dc2b71a3
Add auto-detection for Intel GPUs (#14386)
## Summary

This PR intends to enable `--torch-backend=auto` to detect Intel GPUs
automatically:
- On Linux, detection is performed using the `lspci` command via
`Display controller` id.
- On Windows, ~~detection is done via a `powershell` query to
`Win32_VideoController`~~. Skip support for now—revisit once a better
solution is available.

Currently, Intel GPUs (XPU) do not rely on specific driver or toolkit
versions to distribute different PyTorch wheels.

## Test Plan

<!-- How was it tested? -->
On Linux:

![image](https://github.com/user-attachments/assets/f7f238e3-a797-42ea-b8fa-9b028dfd4db5)
~~On Windows:

![image](https://github.com/user-attachments/assets/a10d774e-1cb9-431b-bb85-e3e8225df98f)~~

---------

Co-authored-by: Charlie Marsh <charlie.r.marsh@gmail.com>
2025-07-09 13:31:08 +00:00
..
accelerator.rs Add auto-detection for Intel GPUs (#14386) 2025-07-09 13:31:08 +00:00
backend.rs Add auto-detection for Intel GPUs (#14386) 2025-07-09 13:31:08 +00:00
lib.rs Automatically infer the PyTorch index via `--torch-backend=auto` (#12070) 2025-03-19 14:37:08 +00:00