Skip to content

Placement solution with scalable overlap loss and deterministic cleanup#35

Open
theSohamTUmbare wants to merge 2 commits intopartcleda:mainfrom
theSohamTUmbare:main
Open

Placement solution with scalable overlap loss and deterministic cleanup#35
theSohamTUmbare wants to merge 2 commits intopartcleda:mainfrom
theSohamTUmbare:main

Conversation

@theSohamTUmbare
Copy link

@theSohamTUmbare theSohamTUmbare commented Dec 28, 2025

Summary

This PR submits my placement solution using:

  • Scalable overlap / overlap-lite loss with spatial hashing
  • Explicit macro–std and macro–macro separation
  • Final deterministic cleanup to guarantee zero overlaps

Results

  • All 10 test cases pass
  • Overlap ratio (all cases): 0.0000
  • Normalized wirelength (largest test case): ~0.19 um
  • Runtime (largest test case): ~33s (CPU)
  • Normalized wirelength of top 10 test cases: 0.2594 um
  • Total Runtime of top 10 test cases: 59.13s

Results on newly added realistic test designs

  • Total overlap ratio: 0.0000
  • Normalized wirelength on 11th test case (10000+ cells): 0.2849 um
  • Total runtime on 11th test case: 222.84s

How to run

python test_placement.py

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant