-
Notifications
You must be signed in to change notification settings - Fork 550
Error Handling: refactor the PjRt registry to use status QOL functions. #9419
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
base: master
Are you sure you want to change the base?
Conversation
3302da5
to
50d9c5a
Compare
This comment was marked as outdated.
This comment was marked as outdated.
9ebf41c
to
8499caf
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks!
a4b9bd4
to
c0e7506
Compare
These CI failures seem to be coming from #9427. |
The CI should be fixed now. Could you rerun the checks and merge? |
Replace `XLA_CHECK` macros with `XLA_ASSIGN_OR_RETURN` and `XLA_RETURN_IF_ERROR` for better error handling: - `InitializePjRt()` now returns `StatusOr<T>` instead of throwing on errors - Enhanced error messages with location context - Consistent error handling across all device types
c0e7506
to
1f0dca3
Compare
I tried rerunning the CI, but the failures are still there. |
@bhavya01 , is this related to the fix you are working on? |
This PR replaces the existing
XLA_CHECK*
macros withXLA_ASSIGN_OR_RETURN
andXLA_RETURN_IF_ERROR
for improved error handling within the PyTorch/XLA codebase. Specifically, theInitializePjRt()
function has been updated to returnStatusOr<T>
.