Just anorth
Just another example of Intel's deceptive practices.
Just another example of Intel's deceptive practices.Update: https://www.tomshardware.com/pc-com...ons-of-cpus-knowing-of-downfall-vulnerability
That complaint PDF should be interesting to read.
Just another example of Intel's deceptive practices.
Yeah, didn't Intel keep selling chips vulnerable to parts or all of SMeltdown after that was discovered? Keep in mind Intel knew about it before we did.That's nothing new though. Clearly Intel's not going to stop producing chips or order a recall because of security flaws that can be "fixed".
Yeah, didn't Intel keep selling chips vulnerable to parts or all of SMeltdown after that was discovered? Keep in mind Intel knew about it before we did.
They couldn't possibly ask for much if the only thing that happened is that a microcode update screwed up game performance.It's obvious that these people want money.
They cannot demonstrate that the slowdowns, if any, happened after the microcode updates incorporating the patches. They all give their subjective opinion, based on stuff like perceived slowdowns in games and photo and video editors, on Windows.They couldn't possibly ask for much if the only thing that happened is that a microcode update screwed up game performance.
I unstickied it, because it was singling out Intel's exploit, when infact we had many from all across, so the title was misleading.Why create a new thread for each new exploit?
Much better to keep it all here.
It is, though the bias is skewed towards one particular company. I mean, a new vulnerability has just been discovered with the products of the 'other' company, but, as usual, the outrage is more in case of the former.Is this the thread where if a company has an identified security threat we're likely to never encounter, we are allowed to bash on that company?
with these "unpredictable results" being -It had found a case where adding redundant rex.r prefixes to an FSRM optimized rep movs operation seemed to cause unpredictable results.
Their analysis of what might be happening is -We observed some very strange behavior while testing. For example, branches to unexpected locations, unconditional branches being ignored and the processor no longer accurately recording the instruction pointer in xsave or call instructions.
and furtherWe believe this bug causes the frontend to miscalculate the size of the movsb instruction, causing subsequent entries in the ROB to be associated with incorrect addresses. When this happens, the CPU enters a confused state that causes the instruction pointer to be miscalculated.
with the most important part being:The machine can eventually recover from this state, perhaps with incorrect intermediate results, but becoming internally consistent again. However, if we cause multiple SMT or SMP cores to enter the state simultaneously, we can cause enough microarchitectural state corruption to force a machine check.
All they have is a hunch that something might be possible!We know that we can corrupt the system state badly enough to cause machine check errors, and we’ve also observed threads interfere with execution of processes scheduled on SMT siblings.
However, we simply don’t know if we can control the corruption precisely enough to achieve privilege escalation. I suspect that it is possible, but we don’t have any way to debug μop execution!
I don't understand how this bug (this is actually a bug that has the potential to cause a security vulnerability) is a result of Intel trying to extend the x86 ISA.
From the blog post:
with these "unpredictable results" being -
Their analysis of what might be happening is -
and further
with the most important part being:
All they have is a hunch that something might be possible!
You can bet that NSA and state actors have that in their arsenal.I suspect that it is possible, but we don’t have any way to debug μop execution!
A messy ISA allows code to put the CPU into an impossible state of confusion that can be used to DoS the system, if not more. How is that not down to the ISA being messy? And don't tell me Intel isn't responsible for the x86 ISA.I don't understand how this bug (this is actually a bug that has the potential to cause a security vulnerability) is a result of Intel trying to extend the x86 ISA.
They did the same thing to the x86 ISA that a sloppy coder does to his or worse, someone else's code. Band aids upon band aids instead of starting from scratch and weeding out legacy crap.And don't tell me Intel isn't responsible for the x86 ISA.
Intel and AMD are looking to increase performance by looking at everything there is to improve upon in their architectures. How does non-standard use of a prefix on an operation that has been enhanced to give better performance with short strings equate to sloppy ISA extension?A messy ISA allows code to put the CPU into an impossible state of confusion that can be used to DoS the system, if not more. How is that not down to the ISA being messy? And don't tell me Intel isn't responsible for the x86 ISA.
See? Without AMD, this new Intel vuln probably would never have been discovered.The irony is that the technique that was used to find this vulnerability in the first place, was originally developed to find a vulnerability in Zen 2