How to create an implementation signoff flow proving that the final FPGA netlist is functionally equivalent to the RTL model.
For standards IEC 61508 / ISO 26262 / EN 50128 / DO-254.
FPGAs are the dominant hardware platform in low-volume, safety-critical applications, including aerospace, and nuclear power plants. Modern FPGAs allow for the implementation of high performance designs with integrated safety mechanisms. This is driving adoption in additional industries, including automotive. Functional safety standards require a rigorous development process to minimize the risk of introducing systematic faults. Some RTL issues may only reveal themselves as bugs in the synthesis netlist. Additionally, synthesis tools manipulate the design to map it into the fixed FPGA structure. These complex transformations present a high risk of introducing bugs. Gate-level simulation and lab testing can only cover a tiny portion of the FPGA functionality, and are likely to miss implementation bugs. Moreover, they are slow to run and hard to debug. This paper presents an implementation signoff flow proving that the final FPGA netlist is functionally equivalent to the RTL model. Based on FPGA-specific, mature formal technology, the solution is exhaustive and efficient, with many issues being caught before synthesis starts.
Read more here.
While terms often are used interchangeably, they are very different technologies with different challenges.
The industry is gaining ground in understanding how aging affects reliability, but more variables make it harder to fix.
Key pivot and innovation points in semiconductor manufacturing.
Tools become more specific for Si/SiGe stacks, 3D NAND, and bonded wafer pairs.
Thinner photoresist layers, line roughness, and stochastic defects add new problems for the angstrom generation of chips.
Less precision equals lower power, but standards are required to make this work.
While terms often are used interchangeably, they are very different technologies with different challenges.
Open-source processor cores are beginning to show up in heterogeneous SoCs and packages.
New applications require a deep understanding of the tradeoffs for different types of DRAM.
Open source by itself doesn’t guarantee security. It still comes down to the fundamentals of design.
How customization, complexity, and geopolitical tensions are upending the global status quo.
127 startups raise $2.6B; data center connectivity, quantum computing, and batteries draw big funding.
The industry is gaining ground in understanding how aging affects reliability, but more variables make it harder to fix.
Leave a Reply