2012 IEEE 25th Computer Security Foundations Symposium 2012
DOI: 10.1109/csf.2012.17
|View full text |Cite
|
Sign up to set email alerts
|

Cache-Leakage Resilient OS Isolation in an Idealized Model of Virtualization

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

0
19
0

Year Published

2013
2013
2023
2023

Publication Types

Select...
8

Relationship

1
7

Authors

Journals

citations
Cited by 27 publications
(19 citation statements)
references
References 38 publications
0
19
0
Order By: Relevance
“…In particular, aws of the virtualisation platform in place, may constitute the guest operating system (OS) vulnerable to side-channel attacks. For example, weak isolation between host and guest OSs may lead to a side-channel attack based on cache leakage [43].…”
Section: Vulnerabilitiymentioning
confidence: 99%
“…In particular, aws of the virtualisation platform in place, may constitute the guest operating system (OS) vulnerable to side-channel attacks. For example, weak isolation between host and guest OSs may lead to a side-channel attack based on cache leakage [43].…”
Section: Vulnerabilitiymentioning
confidence: 99%
“…System security In order to achieve strong isolation, Barthe et al [3] present a model of virtualization which flushes the cache upon switching between guest operating systems. Flushing the cache in such scenarios is common and does not impact the already-costly context-switch.…”
Section: Related Workmentioning
confidence: 99%
“…System security In order to achieve strong isolation, Barthe et al [6] present a model of virtualization which flushes the cache upon switching between guest operating systems. Different from our scenario, flushing the cache in such scenarios is common and does not impact the already-costly context-switch.…”
Section: Related Workmentioning
confidence: 99%
“…For example, modifying an algorithm implementation, as in the case of AES [7], does not naturally generalize to arbitrary untrusted code. Similarly, flushing or disabling the cache when switching protection domains, as suggested in [6,49], is prohibitively expensive in systems like Hails, where context switches occur hundreds of times per second. Finally, relying on specialized hardware, such as partitioned caches [29], which isolate the effects of one partition from code using a different partition, restricts the deployability and scalability of the solution; partitioned caches are not readily available and often cannot be partitioned to an arbitrary security lattice.…”
Section: Introductionmentioning
confidence: 99%