High-level synthesis, verification and language
By John Sanguinetti, CTO of Forte Design Systems Inc.
edadesignline.com (February 22, 2010)
Abstract
The preferred high-level design methodology proceeds from high-level code to RTL code. Good verification practice requires that the input to High-level Synthesis (HLS) be verified first, via simulation (or some other analytical means), and then the output of HLS be verified, again via simulation or some other means. Using SystemC as the input language to HLS enables this flow, but using C as the HLS input language imposes a serious limitation on doing verification this way.
Introduction
High-level design has many advantages over the more commonplace design flow which begins with RTL code. Among the most compelling advantages is the improved verification efficiency which a higher level of abstraction offers. It is apparent to the point of being self-evident that when the source code of a design is created, there will be fewer errors if the source is at a higher abstraction level than if it is at a lower level. However, there is still a process required to verify the transformations which are applied to the design description as it proceeds through the design flow from creation to final realization.
To read the full article, click here
Related Semiconductor IP
- HBM4 PHY IP
- Secure Storage Solution for OTP IP
- Ultra-Low-Power LPDDR3/LPDDR2/DDR3L Combo Subsystem
- MIPI D-PHY and FPD-Link (LVDS) Combinational Transmitter for TSMC 22nm ULP
- VIP for Compute Express Link (CXL)
Related Articles
- How High-Level Synthesis Can Raise the Efficiency of Design Reuse
- High-Level Synthesis - Ready for prime-time?
- The future is High-Level Synthesis
- Building a NAND flash controller with high-level synthesis
Latest Articles
- Making Strong Error-Correcting Codes Work Effectively for HBM in AI Inference
- Sensitivity-Aware Mixed-Precision Quantization for ReRAM-based Computing-in-Memory
- ElfCore: A 28nm Neural Processor Enabling Dynamic Structured Sparse Training and Online Self-Supervised Learning with Activity-Dependent Weight Update
- A 14ns-Latency 9Gb/s 0.44mm² 62pJ/b Short-Blocklength LDPC Decoder ASIC in 22FDX
- Pipeline Stage Resolved Timing Characterization of FPGA and ASIC Implementations of a RISC V Processor