Part numbers of CPU's without the meltdown and spectre bugs?

[imported_ats]

Sure they are correct. They would also be correct in saying:

No, there are multiple chips from multiple vendor using multiple ISA that are vulnerable to at least all three major exploits that have been published.

 

[imported_ats]

It should be obvious intel CPUs are not working as designed or intended. They were intended and designed to not allow lower level processes to access privileged data. They missed something and that means the CPUs have a bug.

Um, they don't allow lower privileged processes to access higher privileged data. If they did, a side channel attack wouldn't be required. This definitely falls into the exploit category and not the bug category.

Spectre as an exploit targets a particular architectural decision (OoO, speculative processing). Which isn't a bug, is a general security issue that needs to be resolved.

Meltdown is an exploit that targets a specific bug or design deficit on Intel processors. Intel CPU's are not doing what they are supposed to be doing and Meltdown takes advantage of that.

Both Meltdown and Spectre are exploits that target speculative execution. Meltdown is not the result of a specific bug OR design deficit. Intel CPUs are doing what they are suppose to due. They are susceptible to a side channel exploit like a wide range of designs from multiple vendors using multiple ISAs.

Yeah. No. The CPUs are not working as intended. The CPUs are intended to prevent code accessing privileged data.

Which they do. Anyone saying they don't is simply lying. Side channel attacks exist specifically because the CPU is preventing code from accessing privileged data, if the CPU wasn't doing the right thing, a side channel attack wouldn't be required.

In regards to Meltdown. It would not exist if Intel CPU's unlike did what it was supposed to when verifying memory access permissions after the code has been processed. It should do a check notice the discrepancy and chuck it. That is what the check is for. That is what all other CPU's do. Whether it was an intentional choice by an engineer when realizing it was doing what it was supposed to do, a design decision by Intel to ignore the security check for performance, or it was just an oversight because when this was originally done, there wasn't really an attack vector that could utilize it anyways, there is a design feature included in the CPU that doesn't handle it's job correctly. That is a bug and not a CPU doing what it is supposed but a new exploit takes advantage of that.

They do do a check and notice the discrepancy and chunk it. They work like almost all other speculative OOO CPUs. There is literally only 1 CPU that does speculative OOO execution and doesn't do it like Intel and that is zen and that wasn't to prevent some bug, that was a power optimization. Power? Same thing as Intel. All other AMD OOO designs? Same thing as Intel. Alpha? Same thing as Intel. MIPS? Same thing as Intel. ARM? Same thing as Intel. Literally every OOO design except Zen has the meltdown issue.

Intel doesn't have any security checks for cache data read tasks.
Intel has security checks in place but the CPU ignores those checks for whatever reason.

These are categorically false statements.

 

[Thala]

ARM? Same thing as Intel. Literally every OOO design except Zen has the meltdown issue.

In ARMs long list of OOO designs, there is only Cortex A75 affected by Meltdown. So your statement is wrong.

In addition, contrary to Spectre, the Meltdown attack is relatively easy to avoid in HW. Part of the reason is, that access privileges are know very early in the data access process. This holds in particular as Meltdown requires an additional data dependency on invalid data in order to leak the information via side channels.
On the other hand, control flow speculation as required by Spectre cannot trivially be avoided without degrading performance, since the evaluation of branch condition happens relatively late.

 

[imported_ats]

In ARMs long list of OOO designs, there is only Cortex A75 affected by Meltdown. So your statement is wrong.

That's not actually true. A15, A57, and A72 are also affected by a variant of Meltdown.

 

[Thala]

That's not actually true. A15, A57, and A72 are also affected by a variant of Meltdown.

Not according to ARM:
https://developer.arm.com/support/security-update

 

[imported_ats]

Not according to ARM:
https://developer.arm.com/support/security-update

You might want to read that document. A15, A57, and A72 are all affected by Variant 3a.

 

[jihe]

80386, 80486

 

[whm1974]

80386, 80486

I know you are kidding, but can you actually do anything with those CPUs in 2018?

 

[WhoBeDaPlaya]

I know you are kidding, but can you actually do anything with those CPUs in 2018?

Retro gaming!
Still have my old AMD DX4-100 rig complete with VESA graphics and caching HDD controller back home

 

[jihe]

No, the CPUs are working as intended, imo. This is not a bug. Someone has just found a way to exploit the way the CPU normally works. That's the way I look at it.
I think we just disagree, and this disagreement is reasonable.
And all CPUs can be exploited.
Some exploits just haven't been discovered yet, or made public yet.

You sound more and more like Apple. Just call it a feature and be done with it.

 

[jihe]

I know you are kidding, but can you actually do anything with those CPUs in 2018?

You can do everything you did in 1990. And for some of us those were the best things in our lives.

 

[2blzd]

Im sure I will get roasted for saying this, but why is everyone so paranoid. A hacker has to specifically target you correct?

 

[Tuna-Fish]

Im sure I will get roasted for saying this, but why is everyone so paranoid. A hacker has to specifically target you correct?

Not really. When operating systems have had flaws of this magnitude in the past, they have typically lead to self-propagating worms that attack the entire installed base.

At least some of these bugs are exploitable from javascript, and with PoC and technical details available, it probably won't be long before the first internet-based attacks where some ransomware or other crap infects any unpatched machines visiting the wrong site through their browser.

 

[Thala]

You might want to read that document. A15, A57, and A72 are all affected by Variant 3a.

Variant 3a is made up by ARM. Please read the Meltdown Whitepaper, where it clearly states that Meltdown is attacking memory content, which variant 3a is not. In fact variant 3a is not even fixable with the common PTI/KAISER patches, which is fixing Meltdown vulnerability. The reason for this is, that Meltdown is attacking mapped kernel pages in user space, while again variant 3a is not.

ARM in their evaluation paper clearly say in reference to variant 3a:

In general, it is not believed that software mitigations for this issue are necessary.

So whatever you name variant 3a, it is not Meltdown.
In particular it is wrong claiming that that A15, A57, A72 are susceptible to Meltdown.

 

[whm1974]

Not really. When operating systems have had flaws of this magnitude in the past, they have typically lead to self-propagating worms that attack the entire installed base.

At least some of these bugs are exploitable from javascript, and with PoC and technical details available, it probably won't be long before the first internet-based attacks where some ransomware or other crap infects any unpatched machines visiting the wrong site through their browser.

Goodbye JavaScript?

 

[richaron]

No, there are multiple chips from multiple vendor using multiple ISA that are vulnerable to at least all three major exploits that have been published.

Lol you are exactly the sort of person who bought into intel's spin but you don't understand my example of spin. Think about it. Nothing I said is wrong, I only referenced what was convenient and handled the scope of my statement. I didn't say it was the only exploit. I didn't say it was the only CPU. I didn't even say it was the only desktop affected by exploits which need to be fixed in software: I said "an exploit" and I am logically correct because I'm referring to something specific. Intel has pulled the same BS with their spin (but obviously to make themselves seem equally affected as everyone else lol)

Once again, it's completely true that:

Intel CPUs are the only desktop and datacentre CPUs affected by an exploit which needs to be fixed in software. This software fix causes up to 30% decrease in performance.

 

[imported_ats]

Im sure I will get roasted for saying this, but why is everyone so paranoid. A hacker has to specifically target you correct?

These exploits are possible via js running in web browsers, fyi.

 

[Tuna-Fish]

Goodbye JavaScript?

According to firefox devs, it's possible to change their JS implementation to protect against this, so there are no future implications for JS.

The problem is all the old machines with crappy old unpatched browsers. Something like 4% of the browser market share is still held by IE.

 

[SPBHM]

According to firefox devs, it's possible to change their JS implementation to protect against this, so there are no future implications for JS.

The problem is all the old machines with crappy old unpatched browsers. Something like 4% of the browser market share is still held by IE.

I think IE11 is still being supported and likely to be patched for this?!

 

[DaveSimmons]

According to firefox devs, it's possible to change their JS implementation to protect against this, so there are no future implications for JS.

The problem is all the old machines with crappy old unpatched browsers. Something like 4% of the browser market share is still held by IE.

Those people are vulnerable to 100+ existing, easier to use attack methods so they are already thoroughly infected without the need for any S&M.

 

[2blzd]

Not really. When operating systems have had flaws of this magnitude in the past, they have typically lead to self-propagating worms that attack the entire installed base.

At least some of these bugs are exploitable from javascript, and with PoC and technical details available, it probably won't be long before the first internet-based attacks where some ransomware or other crap infects any unpatched machines visiting the wrong site through their browser.

These exploits are possible via js running in web browsers, fyi.

Makes sense, thanks for being easy on me