Confronting issues due to UPF instrumentation.
The Unified Power format (UPF) standard enables designers to add power intent for a design. For power management, designers typically partition a design into power domains. Interactions between these power domains are done through various power control logic, like retention logic, isolation logic, and level shifters. Designers need to validate that the power control logic does not introduce new multi-clock and multi-reset issues into the design.
This paper examines issues in reset domain crossings introduced by UPF instrumentation and describes enhancing a static verification tool with new rulesets.
An efficient methodology for power-aware RDC
This paper explains how metastability can be introduced by UPF instrumentation into the reset domain crossing space, describes the challenges designers encounter while working with power-aware RDC analysis, and investigates how newly introduced resets due to UPF instrumentation can affect the data interactions between reset domain paths crossing various power domains. We depict issues in various valid scenarios from a real life SoC and show how low power UPF strategies, like the insertion of isolation cells, can have a huge impact on the reset strategy, as the number of new resets increase dramatically. These issues if undetected, may result in a higher verification turnaround time and unnecessary ECOs, resulting in costly respins.
The conventional RDC methodology still holds good for power aware designs but there is a need to refine the methodology to incorporate certain issues which may creep in due to power instrumentation. We suggest a few RDC checks which are otherwise missing in static verification tools. These checks ensure the correct propagation of resets and reset connectivity to the blocks which are in an on/off power state. We share an efficient methodology to deal with these new power-related RDC issues.
To read more, clickhere.
But gaps in tools make it difficult to address warpage, structural issues, and new materials in multi-die/multi-chiplet designs.
Technology and business issues mean it won’t replace EUV, but photonics, biotech and other markets provide plenty of room for growth.
GlobalFoundries sues IBM; EU’s $47B chips plan; China chip output falls; Space Forge opening US facility; first LPDDR flash memory; transformable nano-scale electronic devices; tabletop 3D X-ray microscopy.
Gate-all-around is set to replace finFET, but it brings its own set of challenges and unknowns.
Checks, balances, and unknowns for AI/ML in semiconductor design.
Less precision equals lower power, but standards are required to make this work.
New applications require a deep understanding of the tradeoffs for different types of DRAM.
Existing tools can be used for RISC-V, but they may not be the most effective or efficient. What else is needed?
The industry is gaining ground in understanding how aging affects reliability, but more variables make it harder to fix.
Technical and business challenges persist, but momentum is building.
The verification of a processor is a lot more complex than a comparably-sized ASIC, and RISC-V processors take this to another layer of complexity.
New memory standard adds significant benefits, but it’s still expensive and complicated to use. That could change.
The industry seems to think it is a real goal for the open instruction set architecture.
Leave a Reply