2018 IEEE International Conference on Cloud Engineering (IC2E) 2018
DOI: 10.1109/ic2e.2018.00025
|View full text |Cite
|
Sign up to set email alerts
|

Scheduling, Isolation, and Cache Allocation: A Side-Channel Defense

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
9
0

Year Published

2019
2019
2024
2024

Publication Types

Select...
5
2

Relationship

0
7

Authors

Journals

citations
Cited by 10 publications
(9 citation statements)
references
References 28 publications
0
9
0
Order By: Relevance
“…Spectre gadget detection can be categorized as static and dynamic detection techniques. One direction of the static analysis technique is to model the Spectre gadget by using its syntax pattern, such as Spectre 1 Scanner from RedHat [5] and MSCV Spectre 1 pass [41], and conduct the pattern search on binaries for potential candidates. These tools produce a large number of false positives.…”
Section: Related Workmentioning
confidence: 99%
See 1 more Smart Citation
“…Spectre gadget detection can be categorized as static and dynamic detection techniques. One direction of the static analysis technique is to model the Spectre gadget by using its syntax pattern, such as Spectre 1 Scanner from RedHat [5] and MSCV Spectre 1 pass [41], and conduct the pattern search on binaries for potential candidates. These tools produce a large number of false positives.…”
Section: Related Workmentioning
confidence: 99%
“…Generally speaking, they try to find a set of potential Spectre gadgets in a program and only patch these gadgets to avoid high runtime overhead. Spectre V1 Scanner from RedHat [5] and MSCV Spectre 1 pass [41] search in binary for gadget patterns, and only patch gadgets that match predefined patterns. Tools like SPECTECTOR [23] and oo7 [43] conduct more advanced static analysis such as symbolic execution and taint analysis to detect Spectre gadgets.…”
Section: Introductionmentioning
confidence: 99%
“…For example, attacks that rely on simultaneous multithreading (SMT) can be thwarted by disabling SMT, which is an increasingly common practice for both cloud providers and end users [10,19,67]. Other approaches propose to partition the shared resource to ensure that its use by the victim cannot be monitored by the attacker [56,64,92,104,118]. For example, Liu et al [64] present a defense to multicore cache attacks that uses Intel CAT [74] to load sensitive victim cache lines in a secure LLC partition where they cannot be evicted by the attacker.…”
Section: Side Channel Defensesmentioning
confidence: 99%
“…Finally, for channels that rely on preemptive scheduling and SMT, one mitigation approach is to erase the victim's footprint from the microarchitectural state across context switches. For example, several works proposed to flush the CPU caches on context switches [17,32,34,35,37,43,44,78,81,92,99,117].…”
Section: Side Channel Defensesmentioning
confidence: 99%
“…The most common target for side-channel attacks is CPU caches, and many of the practical attacks can be prevented by defending this target. There is an extensive body of research in the direction of preventing side channel attacks, ranging from cache isolation [39], to attack detection [19], enforcing non-interrupted execution [32,41], and cache coloring [37]. Yet, they provide only a partial defence as transient execution attacks may use other side-channel targets too [35].…”
Section: Preventing Side Channelsmentioning
confidence: 99%