We've been asked to create a side-channel security online course. You know us. You know what happened next. So, here's the first trailer #spectre#meltdown cc
Today is the release of the final episode of #SideChannelSecurity. The topic is page cache attacks and it will also have a brief recap all of the 30 episodes.
https://youtube.com/watch?v=z1LX4BAz7x8…
We have a few special appearances in there ;)
Our paper "A Security Analysis of CNC Machines in Industry 4.0" has been accepted at DIMVA 2023. Thanks to all my co-authors, Trend Micro and Celada for supporting the research // cc
📢 The #DIMVA23 Call for Tools is up!
https://dimva2023.de/cft/
Send us your proposal by June 1st. We are looking forward to seeing your tools in action!
The program for DIMVA is now available online: https://dimva2023.de/program/
Congratulations to all the authors whose papers have been accepted.
The registration is now also open. Don't miss out on the early bird registration, open until June 10th.
Check out the DIMVA 2023 program at https://dimva2023.de/program/ and join us in Hamburg from July 12 to 14! #dimva2023
DIMVA Tools Arsenal Submissions still possible until June 1st.
Early Registration Deadline: June 10th.
Man kann die #LetzteGeneration für die Wahl ihrer Mittel kritisieren, aber wenn der Staat all die Energie, die er für den Kampf gegen Klimaaktivist*innen aufbringt, in den Kampf gegen die Klimakatastrophe stecken würde, hätten wir kein Problem.
shows that it is possible to steal RSA keys over VM boundaries.
SQUIP exploits the shared μOP scheduler of AMD CPUs to get information about executed instructions on the SMT sibling core.
#SP23
Can we protect the last-level cache against side-channel attacks?
Randomization defenses provide incomplete protection. Partitioning is hard to scale.
Can we use fancy crypto to deliver the best of both worlds in a low-overhead HW defense?
[SassCache, by
Don't you just hate queueing?
Poor CPU instructions are in the Q all the time🥺
Q contention leaks across hyperthreads.
"Are CPU timestamps accurate enough to measure it?"
No, but we can put *them* in the Q and see how quickly they execute!
[SQUIP, by
This year's S&P also has many fine talks for the µarch security aficionado.
Let's try another round of live-tweeting. #SP23@IEEESSP twitter.com/PurnalToon/sta…
about CSI:Rowhammer!
This mitigation completely changes the game and has a lot of implications beyond mitigating Rowhammer. This might become a very influential #SP23 paper.
https://jonasjuffinger.com/papers/csirowhammer.pdf…
Rowhammer is a sneaky suspect: ECC doesn't catch it!
Let's up our game and use a fast cryptographic MAC instead.
"MACs can't correct errors."
Wrong! Well, right, but creative brute-force error correction is practical😎
[CSI: Rowhammer, by Special Agent
Actually, with so much cool research in uarch security at @USENIXSecurity, I’d like to try and livetweet a bit about the talks that are most interesting to me. Let’s see if I can keep up. #usesec22 twitter.com/PurnalToon/sta…
Efficient memory compression is important.
Compression is data-dependent by design. So fast implementations are too.
"Are you pitching timing attacks on... memory compression?"
Yes! Well, Martin is.
[Timing attacks on Memory Compression, by
for your inspiring research and all the mind enriching collaborations: in your group I met some of the smartest and most talented people in the area.
3/N 🧵
I love this paper.
I spent over 20 hours revising a manuscript last week that had positive comments. Reviewer 2 even said “accept this article” as the ONLY comment!
Yet, still 20+ hours and 2 hours navigating the submission portal.
Can your next paper fix the websites…Show more
Ever wondered how long it takes to reformat papers for journal submissions?
We did, and decided to write about it! We calculated time and money lost, interviewed journal editors and researchers, and developed some new guideline propositions.
Read here: https://bmcmedicine.biomedcentral.com/articles/10.1186/s12916-023-02882-y…
Here's the "Der 7. Sinn" Parody from our #ruhrsec talk about the SQUIP side channel: https://youtu.be/xGmdaVScTi4
Here's our @IEEESSP paper on that topic: https://gruss.cc/files/squip.pdf
cc @notbobbytables
am Austrian Computer Science Day 2018 zu verstehen, dass seine Herangehens- und Arbeitsweise stark natur- und weniger ingenieurwissenschaftlich geprägt ist! Anscheinend fordern komplexe Multiskalensysteme das.
Now live at #RuhrSec 2023
"SQUIP or Why We Need to Study Processors Like Nature"
by @notbobbytables and @lavados.
Conference program, more information and details on our website:
https://ruhrsec.de/2023/#itsecurity#itsicherheit#cybersicherheit#conference#cybersecurity…Show more
Significant USENIX Security papers published at least 10 years ago can be considered for the 2023 USENIX Security Test of Time Award. Submit your nomination for this year's award by June 1: https://bit.ly/usesectot#usesec23
is opening an assistant professor position in systems security!
In particular, on the topic of vulnerability detection, analysis, exploitation, and/or remediation. See the full offer here:
https://eurecom.fr/en/job/digital-security…
Feel free to ask me for any questions!
The WOOT papers are finalized! 16 accepted papers out of 30 submissions. Huge congrats to all authors with accepted papers! Student authors, don't forget to apply for Travel Grant (deadline *April 5th*) here: https://ieee-security.org/TC/SP2023/travel_grants.html…🤯 More exciting news soon, stay tuned!
Does this still work? I have awesome news: DRAMSec '23 is happening! https://dramsec.ethz.ch
Here are ten reasons (+ backstory) on why you should submit your work on memory security, get feedback from industry and academia experts, and profit! 👇
I am happy to announce the release of the sev step framework.
It provides functionality similar
to the famous sgx step but for AMD SEV.
While there are still some rough edges, I hope that
it facilitates new attack research on AMD SEV.
Check it out at