收藏切换
Automatic Insertion Method of High-Performance Synchronization Primitives for Ascend Processors
收藏切换
Shuaijiang Li1, 2, 3, Xinyuan Zhang1, 3, Jiacheng Zhao1, 2, 3, *, Xinghui Tian4, Xiyu Shi1, 3, Xiaoxin Xu4, Huimin Cui1, 2, 3
Journal of Computer Research and Development | 2025, 62(8) : 1962 - 1978
Less
收藏切换
Journal of Computer Research and Development | 2025, 62(8): 1962-1978
• Computer Architecture •
Automatic Insertion Method of High-Performance Synchronization Primitives for Ascend Processors
Full
Shuaijiang Li1, 2, 3, Xinyuan Zhang1, 3, Jiacheng Zhao1, 2, 3, *, Xinghui Tian4, Xiyu Shi1, 3, Xiaoxin Xu4, Huimin Cui1, 2, 3
Affiliations
  • 1Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190
  • 2University of Chinese Academy of Sciences, Beijing 100190
  • 3State Key Lab of Processors (Institute of Computing Technology, Chinese Academy of Sciences), Beijing 100190
  • 4Huawei Technologies Co., Ltd., Hangzhou 310051
  • born in 1998. PhD candidate. Student member of CCF. His main research interest includes heterogeneous programming

    born in 1994. Master, engineer. Her main research interest includes heterogeneous compile

    born in 1989. PhD, associate professor. His main research interests include heterogeneous programming, machine learning systems, and compiler infrastructure for domain specific architectures

    born in 1989. Master. His main research interest includes AI compiler

    born in 1996. Master, engineer. Her main research interest includes heterogeneous programming

    born in 1986. Master. His main research interest includes AI system optimization

    born in 1979. PhD, professor. Her main research interests include compiler construction, compiler optimization, and heterogeneous computing

About Author:

born in 1998. PhD candidate. Student member of CCF. His main research interest includes heterogeneous programming

born in 1994. Master, engineer. Her main research interest includes heterogeneous compile

born in 1989. PhD, associate professor. His main research interests include heterogeneous programming, machine learning systems, and compiler infrastructure for domain specific architectures

born in 1989. Master. His main research interest includes AI compiler

born in 1996. Master, engineer. Her main research interest includes heterogeneous programming

born in 1986. Master. His main research interest includes AI system optimization

born in 1979. PhD, professor. Her main research interests include compiler construction, compiler optimization, and heterogeneous computing

Published: 2025-08-01 doi: 10.7544/issn1000-1239.202440093
Outline
收藏切换

Instruction-level parallelism (ILP) is a classic challenge in the field of processor architecture. Domain-specific architectures, such as the Ascend processor, expose more pipeline details to upper-layer software, and compilers/programmers explicitly control the synchronization between pipelines to optimize ILP. However, the physical synchronization resources between pipelines are limited, which limits the improvement of ILP. To address this issue, a high-performance automatic synchronization primitive insertion method for the Ascend processor is proposed. By introducing the abstraction of “virtual synchronization resources”, this method decouples the insertion of synchronization primitives from the selection of physical synchronization resources. Firstly, a heuristic algorithm is proposed to insert virtual synchronization primitives in complex control flow graphs. Then, a significant number of virtual synchronization resources are mapped to an extremely limited number of physical synchronization resources through virtual synchronization primitive merging and other techniques. At the same time, redundant synchronization primitives in the program are removed based on the partial order relationship between instructions, while ensuring program correctness and stringent hardware resource constraints. Experiments on the Ascend 910A platform using instruction-level and operator-level benchmark programs show that the programs with automatically inserted synchronization primitives achieve performance comparable to or on par with those manually inserted by expert programmers, while ensuring correctness.

Ascend processor  /  synchronization primitives  /  heterogeneous programming  /  domain specific architecture  /  automatic insertion
李帅江, 张馨元, 赵家程, 田行辉, 石曦予, 徐晓忻, 崔慧敏. 面向昇腾处理器的高性能同步原语自动插入方法. 计算机研究与发展, 2025 , 62 (8) : 1962 -1978 . DOI: 10.7544/issn1000-1239.202440093
Shuaijiang Li, Xinyuan Zhang, Jiacheng Zhao, Xinghui Tian, Xiyu Shi, Xiaoxin Xu, Huimin Cui. Automatic Insertion Method of High-Performance Synchronization Primitives for Ascend Processors[J]. Journal of Computer Research and Development, 2025 , 62 (8) : 1962 -1978 . DOI: 10.7544/issn1000-1239.202440093
Year 2025 volume 62 Issue 8
144
55
Cite this Article
BibTeX
Article Info
doi: 10.7544/issn1000-1239.202440093
  • Receive Date:2024-02-03
  • Online Date:2026-11-10
  • Published:2025-08-01
Article Data
Affiliations
History
  • Received:2024-02-03
  • Revised:2025-01-20
  • Accepted:2025-03-03
Affiliations
    1Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190
    2University of Chinese Academy of Sciences, Beijing 100190
    3State Key Lab of Processors (Institute of Computing Technology, Chinese Academy of Sciences), Beijing 100190
    4Huawei Technologies Co., Ltd., Hangzhou 310051
References
Share
https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/10.7544/issn1000-1239.202440093
Share to
QR

Scan QR to access full text

Cite this article
BibTeX
Citations
表12种不同金属材料的力学参数
科
Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)
属
Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏
  • BibTeX
  • EndNote
  • RefWorks
  • TxT