CPU bug wars update: the latest attack against AMD CPUs just leaks memory access patterns. This is basically inherent in how caches work. It's not even a speculation attack. Intel still has the lead in designing CPUs that give up all their secrets *by design* in speculation.
-
Show this thread
-
Replying to @marcan42
You think Intel DESIGNED their chips to cause security problems and lose them sales? Come the fuck on.
1 reply 0 retweets 0 likes -
Replying to @blakeyrat
I think intel DESIGNED their chips under the mantra that anything that happens in speculation doesn't matter and no transistor shall be "wasted" on that, and any engineers who brought up security concerns were ignored.
1 reply 0 retweets 2 likes -
Replying to @marcan42
You realize that nobody, nobody anywhere, brought up security concerns about these features until they'd been in every CPU for like over a decade, right? It's not JUST Intel that missed it. Either don't turn an "oops" into "Intel engineers are evil monsters!" Ridiculous.
1 reply 0 retweets 0 likes -
Replying to @blakeyrat
Have you missed the part where Intel has a zillion data leaks and AMD doesn't? There's a difference between "oh we can do interesting things in speculation" and "<Intel> who needs page tables? your virtual address is now physical in speculation".
2 replies 0 retweets 1 like -
Replying to @marcan42 @blakeyrat
The fact of the matter is Intel designed their CPUs under the clearly-flawed assumption that anything that happens in speculation stays in speculation, and with absolutely no concern whatsoever for defense in depth (i.e. stopping bad/illegal things early, not late)
2 replies 0 retweets 1 like -
Replying to @marcan42 @blakeyrat
That stinks of top-down company culture driven by ignorant managers. I'm sure Intel engineers knew about the possibility of things like this happening and were told not to do anything about it (until it became a problem).
1 reply 0 retweets 0 likes -
Replying to @marcan42
Do you have *evidence*? Or are you just "sure"? Via telepathy? If you have evidence, you should release it and get famous in tech circles. I think you're just making shit up, personally.
1 reply 0 retweets 0 likes -
Replying to @blakeyrat
I've had more than one comment from (ex)Intel engineers arrive at me via one path or the other. Some publicly, but I didn't bookmark the URLs; I'm sure you can find some if you look deep enough.
1 reply 0 retweets 0 likes -
Replying to @marcan42
The conspiracy theory goes deeper and deeper! OooOOOOoooOOOO conspiracies! "I have evidence but I can't tell you but if you look it's there but I can't tell you because they came from one path or another but I have evidence. OooOOOOoo"
2 replies 0 retweets 0 likes
Dude, it's not a conspiracy theory, it's shitty management. Happens all the time. It doesn't take a genius to see that the pattern of Intel getting this wrong *repeatedly* and nobody else doing that points at a deeper underlying cultural issue in the company.
-
-
Replying to @marcan42
Why did none of those engineers come forward between designing the CPUs and the first papers demonstrating the security flaws a couple years ago? They had 10 years to do it. They would have been famous in the computer security world.
1 reply 0 retweets 0 likes -
Replying to @blakeyrat
I'm talking about comments like "yeah the culture was going downhill" / "they stopped investing in validation" / "management didn't care about these things", not "oh we all knew every detail about spectre a decade ago!!!1!"
1 reply 0 retweets 0 likes - Show replies
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.