A RISC-V 5-stage pipelined CPU that supports vector instructions. Tape-out with U18 technology.
-
Updated
Dec 2, 2019 - Verilog
A RISC-V 5-stage pipelined CPU that supports vector instructions. Tape-out with U18 technology.
SymbiFlow WIP changes for Verilog to Routing -- Open Source CAD Flow for FPGA Research
Electronic design automation for Minecraft
LangChain based VLSI flow that is able to generate required HDL, testbench, design creation and implementaion scripts.
Open-Source VHDL Synthesis for Alhambra II FPGA Board
SerDes RTL design, verification using UVM and Physical design.
A VHDL and Verilog compiler written from scratch in Rust, with no third-party crates: simulation, synthesis, formal verification, FPGA place-and-route, bitstream generation and JTAG programming.
Lark based DEF file parser
Parameterized binary-tree LUT network verified with a full UVM environment, then carried through an automated RTL-to-GDSII flow to 0-DRC signoff (GPDK045, 250x250 um)
Open-source EDA toolchain: a Rust-based HDL (Silica) compiling to Verilog-2005, plus a full digital design suite — lint, format, synth, sim, place & route, DRC — and a Tauri desktop IDE.
Pipelined odd-even transposition sorter taken through a full RTL-to-GDSII flow (Genus + Innovus, GPDK045): 0 DRC, GDSII sign-off, gate-level SDF verification pre- and post-route
32-bit ALU, 14 ops — full RTL-to-GDS: Cadence Genus synthesis, Innovus P&R, DRC clean on 45nm CMOS
From an AI workload or your own RTL to a chip GDS on IHP SG13G2, checked with DRC and LVS: architecture exploration with every decision explained, X-HEEP SoC or AXI4 IP generation, and an open digital flow from lint and formal to DFT, pad ring and signoff. MCP server.
Full RTL-to-GDSII ASIC implementation of a 100MHz low-power multi-clock digital communication SoC in TSMC 130nm, featuring robust CDC, Integrated Clock Gating (ICG), and complete physical sign-off.
Circuit and system design portfolio - DMT wireline transceiver DSP (14nm FinFET tape-out), on-chip adaptive bit/power loading, multi-lane chip-to-chip interface, RFSoC real-time verification, 28nm back-end flow, and analog/full-custom design
Detailed placement over the OpenDB design database: placement legality checking.
Run physical-design stages and inspect timing for a netlist through an MCP server
32-bit RISC-V ASIC physical design featuring a custom standard cell library, bit-sliced datapath routed in Cadence Virtuoso, and automated Cadence Innovus implementation.
Formalizing place-and-route timing-closure bounds.
To associate your repository with the place-and-route topic, visit your repo's landing page and select "manage topics."