Lucene search

K
attackerkbAttackerKBAKB:555EF2E1-269F-4133-8D13-B67EA80F8CC3
HistoryJun 15, 2020 - 12:00 a.m.

CVE-2020-0543 CROSSTALK

2020-06-1500:00:00
attackerkb.com
27

EPSS

0.001

Percentile

19.4%

Incomplete cleanup from specific special register read operations in some Intel® Processors may allow an authenticated user to potentially enable information disclosure via local access.

Recent assessments:

busterb at June 15, 2020 8:18pm UTC reported:

This continues to bury SGX as an actual security mechanism users should be interested in. For leaking keys where you have local access, this is useful for Intel CPUs manufactured in the last 5 years. For general purpose exploitation though, this is less likely to be useful, and the overall risk of using this mechanism still leaves many developers who might use this feature suspicious as they ever were.

The huge performance degradation of RDRAND also isn’t great, though the real problem is for virtual hosting providers where a malicious process or VM can kill overall memory bus performance. https://www.phoronix.com/scan.php?page=news_item&px=RdRand-3-Percent

There are some funny secret-squirrel uses here for the mitigation, as it enables a totally different side-channel problem, but nothing you’d likely see more as a novelty: <https://twitter.com/Kryptoblog/status/1270601775184334849&gt;

Assessed Attacker Value: 2
Assessed Attacker Value: 2Assessed Attacker Value: 3

References