On Tue 02-04-19 09:44:59, Michal Hocko wrote: > On Tue 02-04-19 15:38:02, Yafang Shao wrote: [...] > > Seems I missed this dicussion. > > Could you pls. give a reference to it? > > The long thread starts here http://lkml.kernel.org/r/nycvar.YFH.7.76.1901051817390.16954@xxxxxxxxxxxxx Thinking about it some more this like falls into the same category as timing attack where you measure the read latency or even watch for major faults. The attacker would destroy the side channel by the read so the attack would be likely impractical. -- Michal Hocko SUSE Labs
