Q1: What are the goals of Synthesis
Convert RTL → gate-level netlist
Maintain functional correctness
Meet timing (setup/hold)
Optimize area and power
Produce PnR-friendly netlist
Output: Timing-clean, area/power optimized, tech-mapped netlist
Q2: What are the Technology Inputs in PnR
Technology-dependent files:
LEF (Std cells, macros, IOs abstract)
Tech LEF (Metal layers, vias, design rules)
Liberty (.lib) (Timing, power)
RC tech files (TLU+, ITF)
Antenna rules
DRC rules
Q3: What are the Design Inputs in PnR
Design-specific files:
Netlist (.v)
SDC (timing constraints)
UPF/CPF (power intent)
DEF (if floorplan exists)
Scan DEF / DFT info
Q4: What are the Types of Cells in PnR
Standard cells
Combinational
Sequential (flops, latches)
Macros
Hard macros
Soft macros
IO cells
Physical-only cells
Fillers
End caps
Well taps
Spare cells
Q5: What are the Types of IO Pads
Input pad
Output pad
Bidirectional pad
Power pad (VDD)
Ground pad (VSS)
Corner pad
Analog / special pads
Q6: What is the Purpose of IO Pads
Interface between chip core and external world
Provide:
Signal buffering
ESD protection
Voltage level compatibility
Q7: What is the Use of Boundary Pad
Protects internal circuitry
Helps maintain:
Proper spacing
ESD safety
Acts as a buffer zone around die
Q8: How Tool Differentiates Std Cell, IO Pad, and Macro
Using LEF attributes:
CLASS CORE → Standard cell
CLASS BLOCK → Macro
CLASS PAD → IO pad
SIZE and FIXED property
Q9: Difference Between Soft and Hard Macro
Feature
Soft Macro
Hard Macro
Shape
Flexible
Fixed
Timing
Estimated
Accurate
Example
Synthesized block
SRAM, PLL
Placement
Tool can change
Fixed
Q10: How Tool Calculates Rectilinear Block Area
Area = Sum of all rectangles
Uses Manhattan geometry
Based on LEF macro definition
Q11: Can We Rotate Macro by 90° or 270°
YES, if allowed in LEF:
SYMMETRY X Y R90
NO, if rotation not defined
Memory macros often allow R0 / R180 only
Q12: 7, 9, 12 Metal Layers — Which Gives More Performance and Why
👉 12 metal layers
Reason:
More upper thick metals
Lower resistance
Less routing congestion
Better clock and power routing
Supports higher frequency designs
Q13: Which Input Files Have Resistance and Capacitance Values
TLU+ files
ITF files
.lib (cell internal RC)
Q14: Why Do We Have Different RC Corners
To model process variations
Different:
Temperature
Metal thickness
Width variation
Examples:
RCmin → Fast routing
RCmax → Worst delay
RCtyp
Q15: How Multi-Cut Via Improves Performance and Yield
Reduces via resistance
Improves current carrying capacity
Enhances electromigration reliability
Improves manufacturing yield
Q16: In Which Stage Normal Flop Converts to Scan Flop
👉 DFT / Scan insertion stage
After synthesis
Before PnR
Q17: Difference Between Normal Flop and Scan Flop
Normal Flop
Scan Flop
1 data input
2 inputs (D + Scan)
Functional mode
Test + functional
No scan enable
Has scan enable
Q18: What is Scan Chain and Where Is It Used
Chain of scan flops connected serially
Used during manufacturing testing
Improves observability and controllability
Q19: Formula for Core, Die, and Std Cell Utilization
Core Utilisation
(Core Std Cell Area / Core Area) × 100
Die Utilization
(Core Area / Die Area) × 100
Std Cell Utilization
(Placed Std Cell Area / Placeable Core Area) × 100
Q20: Formula for Channel Spacing
Channel Width = Routing Demand / Routing Capacity
or practically:
Channel spacing = Track pitch × Number of tracks
Comments
Post a Comment