Correct-by-construction polygon processing capabilities, which together are commonly called a polygon processing engine (PPE), enable a physical designer to perform all forms of layout transformation in a post-stream-out GDS database. In particular, a PPE allows the designer to modify a design to conform to design for manufacturability (DFM) and yield enhancement requirements. Metallization reprocessing, transistor gate poly and active layers adjustments, and contact and via insertion are some of the common applications of a PPE. Measurement of the various topological relationships of the polygons in a physical design, and subsequent sizing of the polygons under the topological constraints, form the basis of a PPE’s operation. In addition, the ability to create simple geometries, such as rectangles, polygons, and circles, helps a physical design verification engineer write scripts that can generate rather complex overlay cells, such as the power and ground plane of a regular multi-routed blocks design or the RDL layer of a flip-chip design, in a correct-by-construction manner.
Two important requirements of an efficient PPE are:
- Its ability to maintain the design hierarchy during the process of layout modification or generation.
- Its ability to make intricate changes in the hierarchical positions of cell instances under various layout configurational constraints.
So, it is often the case that a successful PPE is actually a combination of two general polygon processing capabilities. One group of capabilities is geared more toward polygon measurements and sizing, while the other handles database hierarchy issues.
Reprints
