What problem does it solve?
Setting up a local Chipyard environment with Chisel, sbt, and FIRRTL to generate RTL from hardware configurations is complex and error-prone. Manually staging generated artifacts and creating SRAM mappings for place-and-route adds further overhead to hardware design workflows.
Core Features & Use Cases
- Dockerized RTL Generation: Produce Verilog from Chipyard configs (e.g., TinyRocket) without installing local hardware design toolchains.
- Automated SRAM Mapping: Analyze generated memory modules and map them to sky130hd standard SRAM macros for the target process.
- Artifact Staging: Automatically organize generated Verilog, memory configs, and design specs into the ChipAgent workspace for downstream PnR workflows.
- Use Case: A hardware engineer can generate production-ready RTL for a custom RISC-V core in minutes, with pre-validated memory mappings, and immediately hand off the design to the place-and-route team.
Quick Start
Use the chipyard-docker skill to generate RTL from a Chipyard configuration and stage all required artifacts for the place-and-route workflow.