News for hackers. Stuff that matters.
Re:More Intel speculative execution vulnerabilities by gpm on Wednesday November 13, 2019 @02:13PM (#21528195)
This isn't true unfortunately.

Being able to cause manipulate the control flow of code that already exists on the computer can be sufficient. See netspectre for an example that worked on real google cloud vms and local wired networks.

http://www.misc0110.net/web/files/netspectre.pdf

Re:More Intel speculative execution vulnerabilities by paulddraper on Wednesday November 13, 2019 @02:43PM (#21528595)
Wow, that is impressive.

Yes in theory you could do that, but to actually exploit in practice I would have guessed couldn't be done.

Re:More Intel speculative execution vulnerabilities by icedchai on Wednesday November 13, 2019 @03:43PM (#21529305)
Don't get too excited. From the paper: "In the Google cloud, we leak around 3 bits per hour from another virtual machine." This is, of course, under ideal conditions.
Re:More Intel speculative execution vulnerabilities by paulddraper on Wednesday November 13, 2019 @04:12PM (#21529623)
They estimated that with some dedicated hardware they could improve that by 2-10x.

Still not very useful for an attacker.

But still fascinating and impressive they could do it at all.