Guess we know the answer to this now (thanks to https://www.usenix.org/system/files/conference/usenixsecurity16/sec16_paper_razavi.pdf …) You should disable Memory Deduplication.https://twitter.com/haroonmeer/status/579229248914898944 …
Or fix rowhammer already. WHY is there still no literature on actually fixing it??
-
-
I'm talking about hw or kernel/hv level mitigations, even if expensive like only using 1/2 of ram or perf ctr trapping
-
I think Intel is massively opposed to dealing with it
-
Is there a reason? Denial in hopes they don't get sued and forced to replace every chipset out there?
-
Seems like the existence of viable workarounds would help reduce that risk if anything.
-
Can it be mitigated with conservative timings for the memory controller rather than auto-detected ones?
-
Have you ever seen a list of boards where timings are programmable, tools needed, safe values to set?
-
This is what I mean by complete lack of literature on defense against rowhammer.
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.