Skip to main navigation Skip to search Skip to main content

Eternal war in memory

  • Laszlo Szekeres
  • , Mathias Payer
  • , Lenx Tao Wei
  • , R. Sekar
  • Stony Brook University
  • University of California at Berkeley
  • FireEye, Inc.

Research output: Contribution to journalArticlepeer-review

30 Scopus citations

Abstract

Software written in low-level languages like C or C++ is prone to memory corruption bugs that allow attackers to access machines, extract information, and install malware. The war in memory is fought by researchers developing defense mechanisms and attackers finding new ways around these protections. Researchers have developed defense mechanisms protecting applications from different forms of attacks. People first need to understand the attack process in order to analyze and compare protection mechanisms. Most often, attackers exploit memory corruption to control program execution by diverting its control flow. Control-flow hijack attacks use memory errors to corrupt a code pointer. Memory Safety mitigates memory corruption by preventing both spatial and temporal errors. Type-safe languages enforce this policy by disallowing pointer arithmetic, checking object bounds at array accesses, and using automatic garbage collection instead of manual memory management.

Original languageEnglish
Article number6824529
Pages (from-to)45-53
Number of pages9
JournalIEEE Security and Privacy
Volume12
Issue number3
DOIs
StatePublished - 2014

Fingerprint

Dive into the research topics of 'Eternal war in memory'. Together they form a unique fingerprint.

Cite this