An Agile Verification for Multi-core Processor Off-chip Memory System

Multi-core processors utilize off-chip DDR memory to provide high-bandwidth and large-capacity memory access. The off-chip memory system is crucial to processor chip research and development. Seventy percent of the research and development cycle is the verification time, because DDR memory design parameters and working modes are diverse, the verification is time-consuming and resource-consuming, and its verification often becomes the bottleneck of the entire research and development cycle. This paper presents an agile verification platform for off-chip memory systems for multi-core processors and introduces key technologies for improving the efficiency of agile verification. The proposed verification platform includes high-quality test programs and an optimized verification procedure. The test program uses directed and random tests based on prior knowledge and NSA (The Negative Selection Algorithm, NSA), respectively, and three acceleration strategies are designed for the initialization, debugging, and running phase of the verification procedure. To validate the effectiveness of the self-designed 16-core processor, the RTL code of an actual off-chip memory system is used as the verification object. The experimental results demonstrate that when the random test program case is reduced by 40%, there is a 29% improvement in the functional coverage and a 3 to 20 times improvement in verification efficiency. Additionally, a total of 26 design bugs are discovered across four categories. The verification results clearly indicate that the proposed method enhances verification efficiency and is ideal for full-chip functional test and regression test.

Authors

Institutions

Publication Details

Published
2024-09-20
DOI
https://doi.org/10.1109/iccs62517.2024.10846301
Primary Topic
Embedded Systems Design Techniques
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

An Agile Verification for Multi-core Processor Off-chip Memory System

Guoteng Pan, Junbo Tie, Hongjun He, Ying Zhang et al.
Embedded Systems Design Techniques
article

An Agile Verification for Multi-core Processor Off-chip Memory System

Guoteng Pan, Junbo Tie, Hongjun He, Ying Zhang, Zhou Li, Li Luo
article en

Abstract

Multi-core processors utilize off-chip DDR memory to provide high-bandwidth and large-capacity memory access. The off-chip memory system is crucial to processor chip research and development. Seventy percent of the research and development cycle is the verification time, because DDR memory design parameters and working modes are diverse, the verification is time-consuming and resource-consuming, and its verification often becomes the bottleneck of the entire research and development cycle. This paper presents an agile verification platform for off-chip memory systems for multi-core processors and introduces key technologies for improving the efficiency of agile verification. The proposed verification platform includes high-quality test programs and an optimized verification procedure. The test program uses directed and random tests based on prior knowledge and NSA (The Negative Selection Algorithm, NSA), respectively, and three acceleration strategies are designed for the initialization, debugging, and running phase of the verification procedure. To validate the effectiveness of the self-designed 16-core processor, the RTL code of an actual off-chip memory system is used as the verification object. The experimental results demonstrate that when the random test program case is reduced by 40%, there is a 29% improvement in the functional coverage and a 3 to 20 times improvement in verification efficiency. Additionally, a total of 26 design bugs are discovered across four categories. The verification results clearly indicate that the proposed method enhances verification efficiency and is ideal for full-chip functional test and regression test.

National University of Defense Technology (CN)
National Natural Science Foundation of China
Affordable and clean energy
Openalex Percentile: Top 71%
Embedded Systems Design Techniques
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.