Power/Ground Grid Netlist Generator

Generates an ASIC PG grid model from YAML + ITF inputs:

1. output/pg_grid_netlist.sp (SPICE RC + instances + .MEAS)
2. output/pg_grid_visualization.html (2D grid + stack cross-section)
3. output/pg_grid_summary.txt (ASCII report)

Links

• Live demo: https://smprather.github.io/pg-grid-netlist-gen
• Configuration documentation: docs/configuration.md

Requirements

• Python >=3.14
• uv (recommended)

Setup

Linux/macOS

uv pip install -e .

Windows

uv pip install -e .

Run

Linux/macOS

./pg_grid_netlist_gen config.yaml

Windows

.\.venv\Scripts\pg-grid-netlist-gen.exe config.yaml

Or use the batch script wrapper:

.\pg-grid-netlist-gen.bat config.yaml

Common options (all platforms):

./pg_grid_netlist_gen config.yaml --open-browser
./pg_grid_netlist_gen config.yaml --output-dir output

Or on Windows:

.\.venv\Scripts\pg-grid-netlist-gen.exe config.yaml --open-browser
.\.venv\Scripts\pg-grid-netlist-gen.exe config.yaml --output-dir output

Configuration

Primary inputs:

• config.yaml
• ITF file referenced by itf.file (for example freepdk3_rctyp.itf)

Highlights:

• Grid layer usage supports stripe (grid) and staple (staple) layers.
• Metal resistance uses ITF RPSQ.
• Via resistance uses ITF RPV.
• Standard-cell PG pins tap directly onto lowest metal grid nodes by pin type (power / ground), not by pin name.
• IR-drop measurements are generated for both output rise and output fall per instance, with edge mapping based on standard_cells[].unateness.
• Optional external cell model include: standard_cells[].spice_netlist_file.
• Netlist includes HSPICE probing for instance pin voltages via .PROBE V(X*).

Output Regeneration Hook

This repo includes a versioned pre-commit hook at:

• .githooks/pre-commit

It regenerates outputs from config.yaml and stages output/ artifacts before each commit.

Enable it once per clone:

git config core.hooksPath .githooks

Notes

• Default outputs are written under output/.
• The HTML visualization uses legend toggles for layers (no dropdown selector).
• The cross-section plot is included below the 2D grid view.