Two Pass Synthesis

As we have seen in the previous floorplanning post - Steps of Floorplanning and its output, we saw a new term named Two Pass Synthesis. Today, we are talked about it.

 

As the technology moves towards deep submicron or ultra-deep submicron, the accurate analysis of design become important in terms of timing, delay calculation and some other parameter. Among all, interconnect parasitic has become the plays an important role in path delay calculation.

 

What is the Two pass synthesis?

The process of getting new optimized netlist for placement from synthesis is called Two Pass Synthesis. Once floorplan of the block is ready for placement, at that stage, we have to perform such synthesis.

So, what happens?

Initial synthesis with defined or default physical constraints is performed during the first pass in order to generate the initial netlist. Synthesis with actual floorplan is performed during the second pass.

During the second pass, it uses floorplan physical constraints to create an optimized netlist. The main reason for using floorplan physical constraints is to accurately estimate interconnect parasitic and improve timing by considering floorplanning information during optimization.

For two pass synthesis, we need physical constraints information such as

• Die area
• Core area
• Shape of a block
• Port location
• Macro location and orientation
• Keepout margin
• Placement blockage
• Pre-routes
• Bounds 
• Vias
• Tracks
• Voltage area 
• Wiring keepout margin

If timing QoR is not critical, the re-synthesis can be skipped. Once we get optimized netlist, we can use new netlist directly for placement.

 

At this note of discussion, we wrapped up this topic and we will meet soon with a new topic of Floorplan. Thank you. Have a nice day!!!