Lightweight and Secure Branch Predictors against Spectre Attacks

2022 27th Asia and South Pacific Design Automation Conference (ASP-DAC) Pub Date : 2022-01-17 DOI:10.1109/asp-dac52403.2022.9712481

Congcong Chen, Chaoqun Shen, Jiliang Zhang

引用次数: 1

Abstract

Spectre attacks endanger most of CPUs, operating systems and cloud services due to the sharing of branch predic- tors in modern processors, while existing defenses fail to balance the security and overhead. This paper designs a lightweight and secure branch predictor (LS-BP), which provides lightweight hardware isolation for different branch entries of same-address- space and cross-address-space. Therefore, it is difficult for the attacker to establish branch conflicts. Experimental results show the average performance overhead is less than 3% while providing strong protection.

查看原文本刊更多论文

针对幽灵攻击的轻量级安全分支预测器

由于现代处理器中分支谓词的共享，Spectre攻击危及大多数cpu、操作系统和云服务，而现有的防御措施无法平衡安全性和开销。本文设计了一种轻量级、安全的分支预测器(LS-BP)，为同一地址空间和跨地址空间的不同分支项提供轻量级的硬件隔离。因此，攻击者很难建立分支冲突。实验结果表明，在提供较强保护的同时，平均性能开销小于3%。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

2022 27th Asia and South Pacific Design Automation Conference (ASP-DAC)

自引率

0.00%

发文量