Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Function verification structure for artificial intelligence processor chip

A technology of artificial intelligence and functional verification, applied in the direction of electrical digital data processing, special data processing applications, instruments, etc., can solve the problems of less consideration of vector and tensor support, lack of interconnection of artificial intelligence software models, etc., to save production resources, shorten the verification convergence cycle, and reduce the effect of development complexity

Active Publication Date: 2019-08-30
上海芷锐电子科技有限公司
View PDF11 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004]The current mainstream chip function verification structure such as the UVM unified verification methodology platform is based on the System Verilog (SV) language, which was produced before the rise of artificial intelligence chips, focusing on Fine-grained control at the physical level of the chip, less consideration of support for vectors and tensors, and lack of interconnection with artificial intelligence software models

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
  • Function verification structure for artificial intelligence processor chip
  • Function verification structure for artificial intelligence processor chip
  • Function verification structure for artificial intelligence processor chip

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] The present invention is described in further detail now in conjunction with accompanying drawing.

[0036] Such as figure 1 and Figure 8 As shown, a function verification structure for an artificial intelligence processor chip disclosed in the present invention includes a vector listener, a domain duplex transceiver, a vector comparator, and an interface formatting mode.

[0037] Such as Figure 8 As shown, in the vector listener and vector package generation integration process, the vector listener is connected to the interface of the artificial intelligence processor function module by the Bind connection module. The vector listener, vector package, and Bind connection modules are all based on the script program Figure 5The interface formatting pattern shown is automatically generated; the vector listener is automatically connected to the source code port of the artificial intelligence processor function module by the Bind connection module, which can significan...

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

The invention discloses a function verification structure for an artificial intelligence processor chip. The function verification structure comprises a vector listener, a domain duplex transceiver, avector comparator and an interface formatting mode. The interface formatting mode automatically generates a vector packet and a vector listener after the interface formatting mode is processed by a scripting language, and automatically integrates the vector listener into a function verification structure. The vector listener collects an input / output interface vector packet of the artificial intelligence processor function module. The domain duplex transceiver sends an input interface vector packet of the processor function module to an artificial intelligence software model control domain, and returns a processing result to the vector comparator. The vector comparator compares and judges an output vector packet of the processor function module and an output vector packet processed by theartificial intelligence software model control domain. According to the structure, the complexity of collaborative development of artificial intelligence software and hardware is reduced, the development efficiency of the artificial intelligence processor chip is improved, and the convergence period and the chip research and development period of function verification of the artificial intelligence processor chip are remarkably shortened.

Description

technical field [0001] The invention belongs to the field of artificial intelligence, and in particular relates to a function verification structure for an artificial intelligence processor chip. Background technique [0002] Chip functional verification means that the verification team builds a suitable verification platform based on the functional design description of the chip, finds out the lack of functions and performance defects before chip production, and assists the chip design team to test its repair plan. High-quality and efficient functional verification is an important guarantee to ensure that chips are produced on time and with high quality. [0003] Artificial intelligence processor chips need to process massive amounts of vector data, need to support continuously developing deep learning networks such as Vggnet, Resnet, and Yolo, and need to be developed collaboratively with artificial intelligence software, all of which pose higher challenges to the chip fun...

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
IPC IPC(8): G06F17/50
CPCG06F30/398
Inventor 史佳欢刘红亮王斌
Owner 上海芷锐电子科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products