Overview
Optimized for efficient hardware implementation, the Synopsys Foundation Cores include a library of mathematical and floating point (FP) components that allow designers to make tradeoffs in power, performance and area to control design precision and meet design requirements. The new flexible floating point (FFP) format enables designers to make tradeoffs in accuracy and share common operations.
The components library includes a robust set of atomic operators, compound operations and components that can share stages of the FP operators based on a FFP format.
Using the FFP format, designers can implement their own specialized FP components. In particular, the FFP format enables trading off accuracy for better QoR for designs that combine multiple FP operations. This allows designers to explore the area and accuracy of the components to meet their design-specific requirements.
Learn more about FPU IP core
Is your NPU DOOMed? Quadric's Chimera GPNPU runs every AI model — and a complete DOOM engine. Find out why Quadric is different.
In this article, we go through a formal-based, easy-to-deploy RISC-V processor verification application. We show how, together with a RISC-V ISA golden model and RISC-V compliance automatically generated checks, we can efficiently target bugs that would be out of reach for simulation.
Processor verification, however, is never trivial but requires combining the strengths of multiple verification techniques. This technical paper considers how to efficiently verify a RISC-V processor using a multi-layered approach known as the Swiss cheese model adapted from the world of avionics.
Paula Jones, Cadence