Replies: 2 comments 1 reply
-
That seems like a bug. Is it possible for you to package a test case and open an issue? |
Beta Was this translation helpful? Give feedback.
1 reply
-
If you are using ORFS then the issue creation page gives some guidance
|
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
I have successfully run the complete flow for the Black Parrot design using the following command:
After executing the complete flow, I am interested in examining the changes in timing metrics at different stages of the design process. To that end, I have executed placement with:
For reporting timing metrics, I appended both global_place.tcl and resize.tcl scripts with the following commands:
As a result, I obtained the following metrics:
After global placement:
And after resizing:
Given that there is a negative slack, I am puzzled as to why the TNS would return a result of 0. Could this be an issue of precision rounding?
Furthermore, if I aim to optimize the existing macro placement and global placement algorithms specifically for timing metrics, at which step in the entire OpenROAD process do you believe it would be most valuable to consider report_timing? For instance, should I look at it post-CTS, post-detailed routing, or after resizing? Reporting early might show very poor timing metrics; however, after CTS or detailed routing, the timing may meet the requirements, making it difficult to discern differences between algorithms.
Thank you very much for your time and assistance!
Beta Was this translation helpful? Give feedback.
All reactions