Architecture guided optimal system precision definition algorithm for custom integrated circuit

An integrated circuit, the best technology, applied in the field of automatic selection of fixed-point expression method or floating-point expression method, can solve the problem of sub-optimal expression format, system designer can't control exponent or mantissa, etc., to achieve power and The effect of improved performance characteristics

Inactive Publication Date: 2013-09-18
ALGOTOCHIP
View PDF4 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In cases where a floating-point representation different from the model is the form of choice, the system designer has no control over the bits used for the exponent or mantissa and has to deal with sub-optimal representation formats

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Architecture guided optimal system precision definition algorithm for custom integrated circuit
  • Architecture guided optimal system precision definition algorithm for custom integrated circuit
  • Architecture guided optimal system precision definition algorithm for custom integrated circuit

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0018] Figure 2A Example system operations for automatically generating custom ICs with automatic digital representation format selection are shown. The system automatically designs and generates custom integrated circuits based on an algorithmic process or code as input and using highly automated tools that require little human involvement.

[0019] now turn to reference Figure 2A , the process receives and analyzes an algorithm (10) specified in a high-level language such as C or Matlab model. In one embodiment, it is assumed that a first version of the product is developed using floating point representation. This could be any type of floating point representation but typical embodiments may use 32 bit IEEE floating point representation. The process entails identifying one or more of the following with the product's developer or system architecture.

[0020] Parameters in floating point format to optimize.

[0021] Working C code or model of the system.

[0022] Rea...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

Systems and methods are disclosed to automatically determine an optimal number format representation for a model or code to be implemented in a custom integrated circuit (IC) by determining a ratio of dynamic range to static range in the model or code, and selecting a floating point or a fixed point number representation based on the ratio; determining the optimal number representation format based on a cost function that includes hardware area and power cost associated with a predetermined bit precision arithmetic; automatically generating a processor architecture customized to the optimal number representation format; and synthesizing the generated processor architecture into a computer readable description of the custom integrated circuit for semiconductor fabrication.

Description

technical field [0001] The present invention relates to a method for automatic selection of fixed-point representation or floating-point representation in a custom integrated circuit (IC) or application specific integrated circuit (ASIC). Background technique [0002] As electronic devices have become more user-friendly, features added to these electronic devices have proliferated. For example, in a typical hardware product development cycle for embedded applications, algorithm specialists begin with algorithm development. During implementation, engineers translate the design into hardware using ASIC solutions or using programmable solutions such as digital signal processor (DSP) cores. Where this is possible, the designer uses tools provided by the DSP vendor to develop assembly code to target the DSP core. Because performance / memory footprint is so critical, firmware needs to be developed with engineers who understand the algorithm as well as the microarchitecture and to...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06F17/50
CPCG06F30/30G06F2117/08G06F17/10G06F7/483
Inventor 阿南德·潘德伦根皮尔斯·吴西瓦·塞法拉萨蒂许·帕德马纳班
Owner ALGOTOCHIP
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products