News for hackers. Stuff that matters.

A comment on Comparing DDR5 Memory from Micron, Samsung, SK Hynix. Parent comment

Re:Comparing DDR5 Memory from Micron, Samsung, SK Hynix by Dylan16807 on Tuesday February 15, 2022 @09:39PM (#30355742)
You don't fix rowhammer with ECC, you fix it by tracking row activation counts and then refreshing nearby rows. Without any cheap-outs like using small hash tables.
[ Parent | on HN ]
Re:Comparing DDR5 Memory from Micron, Samsung, SK Hynix by notriddle on Tuesday February 15, 2022 @10:10PM (#30355930)
https://comsec.ethz.ch/research/dram/blacksmith/
[ Parent | on HN ]
Re:Comparing DDR5 Memory from Micron, Samsung, SK Hynix by Dylan16807 on Tuesday February 15, 2022 @10:20PM (#30355995)
I've read that paper, yes. It works by bypassing target row refresh, because all those chips are trying to be clever about their row tracking and failing.

Track row activation counts for real, no estimates and no hash tables, and you're not vulnerable to this.

[ Parent | on HN ]