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.

Sign up for GitHub

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

Jump to bottom

CPU fallback limitations #206

Open
fengyuan14 opened this issue May 13, 2024 · 0 comments
Open

CPU fallback limitations #206

fengyuan14 opened this issue May 13, 2024 · 0 comments
Labels
documentation Improvements or additions to documentation
Milestone
PT2.7

Comments

@fengyuan14
Copy link
Contributor

fengyuan14 commented May 13, 2024 •
edited
Loading

📚 The doc issue

Record limitations of CPU fallback during development. These will be the reference/check-list when we get a bug of CPU fallback in future.

  1. View/Tensor meta modification operators cannot be fallback to CPU. Breaking semantics. Requiring change Tensor metas inplace.
  2. Tensor factory operators cannot be fallback to CPU. Backend specific implementation.
  3. RNG operators cannot be fallback to CPU. Backend specific implementation.
  4. Fallback of compound operators don't work. The priority of compound operator dispatch is higher than fallback.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
documentation Improvements or additions to documentation
Projects
None yet
Milestone
PT2.7
Development

No branches or pull requests
3 participants
@chuanqi129 @fengyuan14 @riverliuintel