Kruti Deepan Panda

Case study

Taking PicoRV32 through OpenLane.

An academic exploration of processor configuration, physical implementation, and timing tradeoffs.

← Academic projects

When
January – April 2022
Where
NITK · Prof. Ramesh Kini
Team
Kruti Deepan Panda and Rahul Magesh

The question

How do processor configuration choices translate into physical-design constraints? Our EC383 project explored PicoRV32 configurations through OpenLane using SkyWater 130nm technology.

Architecture and implementation

We compared PCPI-enabled configurations, AXI interfaces, multiplication, and two-cycle ALU and comparison options. Routing memory demands led us to focus on the base core with PCPI. Our contribution was configuration exploration and implementation-flow tuning; PicoRV32 and OpenLane are upstream projects.

OpenLane implementation flowPicoRV32 configuration passes through synthesis, floorplanning, placement, clock tree synthesis, and routing to GDSII. Timing reports guide configuration changes.PicoRV32 configurationSynthesisFloorplanning and placementClock tree and routingTiming review and GDSII
Configuration choices meet physical constraints through the implementation flow.

Decisions and verification

We adjusted synthesis strategy, density, and clock-related settings while investigating timing and routing issues.

Implementation results

The documented runs, reports, and layout views trace the configuration changes through the OpenLane flow. This work covers implementation-flow exploration and does not include fabricated-silicon measurements.

Project files

Repository and implementation notes

Project report (PDF)