mpl: make cluster placement failure message more user-actionable - #11148
Conversation
Signed-off-by: Arthur Koucher <arthurkoucher@precisioninno.com>
|
@jeffng-or FYI. |
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request refactors the error logging in HierRTLMP::placeChildren when the annealing engine fails to find a valid solution, providing more specific error messages depending on whether the failure occurred at the root cluster or a sub-cluster. The review feedback correctly identifies a potential null pointer dereference on best_sa if the logger does not terminate execution, and suggests adding an explicit return; statement to ensure robust error handling.
joaomai
left a comment
joaomai
left a comment
There was a problem hiding this comment.
I think it would be nice to also report the tried utilization values from the utilization list to better guide the user. If it failed, it tried all values and none worked right?
|
@joaomai I agree. I actually think it would be better to have a more informative output throughout the execution so that it's clear that the utilization mutates based on the amount of halo area. How about we merge this change and I'll look into more descriptive logging subsequently? |
2 participants
No description provided.