datasheets.com EBN.com EDN.com EETimes.com Embedded.com PlanetAnalog.com TechOnline.com  
Events
UBM Tech
UBM Tech

Design Article

Comment


OphirT

11/28/2012 3:00 AM EST

The first big difference is that at the end of the suggested flow you have ...

More...



WiLess

11/26/2012 2:08 PM EST

An interesting idea and really helpful. What would be the differences of this ...

More...

Speeding power estimation from weeks to hours

Ophir Turbovich, Cambridge Silicon Radio and Thomas Li, SpringSoft

11/19/2012 10:59 AM EST

Introduction

This paper describes a new methodology that automatically generates a chip design’s gate-level waveform from the RTL design environment without the need to bring up the gate-level environment. The new waveform generation methodology reduces the effort to perform gate-level power estimation from weeks to hours, using established EDA technology from Springsoft  and Cambridge Silicon Radio's established power estimation flow and tools. This major reduction in effort and increase in designer productivity enables CSR to analyze power characteristics much earlier in the design flow than is practically possible using traditional, high-effort gate-level analysis. Moreover, the new methodology produces waveforms identical (or nearly identical) to those generated by gate-level simulation. Consequently, the design can be analyzed and optimized iteratively throughout the post-synthesis design flow, enabling much earlier detection and easier resolution of power issues. The paper discusses:

  • Power analysis challenge
  • New automated gate-level waveform methodology
  • Springsoft’s Siloti™ Visibility Automation System
  • Analysis results

Power analysis challenge

One approach to reducing power analysis effort is to perform it at the register transfer level (RTL). Although faster and easier than gate-level analysis, its accuracy is limited to about 20 to 25 percent because synthesis/place and route are subject to many variables. These variables include the synthesizer’s various approaches to meeting timing constraints, as well as its RTL implementation choices, for example, the use of special cells or clock tree insertion techniques. Consequently, RTL analysis is inadequate for the fine-tuning necessary to achieve the low power goals of a complex VLSI design. It is suitable for comparing the power characteristics of design revisions and tracking trends. However, it is certainly not accurate enough to sign-off the design to production, especially a design that must meet a stringent power specification.

There are several approaches to estimating power consumption accurately. One of the common approaches is to estimate it on the post-place and route netlist using full annotation generated from simulation of real scenarios on the same netlist, with or without the standard delay format (SDF). In this approach all the clock tree and all the wire capacitance are taking into account in the most accurate way. This approach necessitates bringing up the gate-level design environment.





WiLess

11/26/2012 2:08 PM EST

An interesting idea and really helpful. What would be the differences of this flow compared to the flow when RTL VCD along with gate-level netlist is loaded into Primetime PX and tool propagates RTL activities into gate nodes?

Sign in to Reply



OphirT

11/28/2012 3:00 AM EST

The first big difference is that at the end of the suggested flow you have waveform file that you can use for other needs too (and other tools too)…
Also – while working with PTPX with RTL VCD you must make sure that only Flip-Flops are mapped between RTL to netlist. Usually more than Flip-Flops are mapped, and since the RTL simulation is zero delay simulation, the propagation is not done well (and the results are not accurate). It is mostly critical and problematic around adders and multipliers, but not only.
The suggested flow can overcome partial design – RTL which contain stubs or other testbench that replace part of the RTL. You can’t overcome this in PTPX.
In general – you must be PTPX expert to be able to run with RTL VCD, and it is not accurate. In the SpringSoft flow anyone can generate netlist VCD/SAIF, and no need to be an expert to run it on PTPX.
Another issue – PTPX runs in this mode taking too much time. While generating GTL waveform in the new flow took about 1 hour, and the total power estimation was 2 hours, using this on PTPX (when it can be done) took more than 24 hours.

Sign in to Reply



Please sign in to post comment

Navigate to related information

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)