Invention Grant
- Patent Title: Fast and efficient system and method for detecting and predicting rowhammer attacks
-
Application No.: US17357265Application Date: 2021-06-24
-
Publication No.: US11887686B2Publication Date: 2024-01-30
- Inventor: Yier Jin , Yichen Jiang , Xuan Zhang , Huifeng Zhu , Xiaolong Guo
- Applicant: University of Florida Research Foundation, Incorporated , Washington University
- Applicant Address: US FL Gainesville
- Assignee: University of Florida Research Foundation, Incorporated,Washington University
- Current Assignee: University of Florida Research Foundation, Incorporated,Washington University
- Current Assignee Address: US FL Gainesville; US MI St. Louis
- Agency: Alston & Bird LLP
- Main IPC: G06F21/57
- IPC: G06F21/57 ; G11C11/406 ; G11C11/4076 ; G11C29/52 ; G11C11/4096 ; G11C7/24 ; G11C11/4078 ; G11C29/04 ; G11C29/20

Abstract:
Embodiments provide for predicting rowhammer attack vulnerability of one or more memory cells of a direct random access memory (DRAM) chip, the DRAM chip including a plurality of memory cells. An example method, determines, for each memory cell of a subset of memory cells of the plurality of memory cells, a leakage time t, a resistance of intrinsic leakage RL based at least in part on the leakage time t, an activation time of an adjacent aggressor row to flip a bit in the memory cell, a resistance of coupling leaking RSW based at least in part on the activation time, and a toggling count. The method identifies, based at least in part on one or more of the RSW, RL, or toggling count, whether the direct random memory access (DRAM) chip is vulnerable to a rowhammer attack.
Public/Granted literature
- US20210406384A1 FAST AND EFFICIENT SYSTEM AND METHOD FOR DETECTING AND PREDICTING ROWHAMMER ATTACKS Public/Granted day:2021-12-30
Information query