Sidebar Sidebar
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

Intel Claims Security Flaw Also Impacts Non-Intel Chips, Exploits Can't Corrupt, Modify or Delete Data [Updated]

Hodar1 said:
That statement WAS true back in the 32 bit days. AMD/Intel could share the same socket on motherboards, and often did. Then the 486DX fiasco came out, where Intel copyrighted the socket footprint, and the AMD/Intel "partnership" effectively shut things down. They still continued to use the same chipset for years afterwards, just different pin-out on the CPU.
Fast forward to 64 bit days, and the bus architecture Intel uses is in stark difference to the HyperTransport system that AMD, NVidia, Apple, Broadcom, SGI, Sun and others all belong to (www.hypertransport.com). Today, AMD's design efforts seek to "mimic" Intel in operations, but seek to utilize alternative design approaches internally. Therefore, having worked there years ago in an engineering role; I disagree with the AMD "just copy Intel" statement. They seek to be compatible with Intel operations, but with Ryzen the internals are functionally different. If this exploit still exists, for me, is still an unsettled case.
Click to expand...
Also the x64 operations in the Intel chipset belong to AMD. They agreed to share certain technology and cut out any CPU new comers (Nvidia!).
 
  • Like
Reactions: fairuz
I read Intel CPUs since 1995 except Atom until 2013 and Itanic are affected by the vulnerability of which AMD is free.
 
GeminiCricket said:
From AMD's statement:
"AMD is not susceptible to all three variants."
If written by a native English speaker, this is a VERY curious wording. Does this really mean that AMD is not susceptible to ANY of the three variants, or does it mean that AMD might be susceptible to one or two of the variants?
Click to expand...
I also noticed that. I'm guessing that it is a well thought out statement hoping that most people will not see the nuance. I take it to mean that AMD is affected by some of the variants.
 
  • Like
Reactions: nt5672
Jensend said:
He sold his shares at 44.555

Intel Corporation (INTC) closed at 45.26 today.
Click to expand...
But he couldn't predict that and you cannot stop and wonder if he didn't sell all his shares (keeping the CEO minimum) to not risk loosing a lot of money.
Also worth noting that with the amount of shares he held he would have pocketed a quarter million dollars a year in dividends. So unless he's short for cash, not sure why he would sell now. Specially what the price target for Intel's stock is.
I usually don't like the fool.com but the article they posted has a lot of good information.
 
  • Like
Reactions: bowmasters
Okay, now that the actual security details are out, finally... it may very well turn out to affect AMD and every other processor maker in existence. The CERT advisory lists AMD as "affected". And given that it's a timing attack, and AMD chips do speculative execution (as does every modern processor) - it's way to early too make the claim it's Intel-only.

Indeed the security researchers simply state they've only tested it on Intel. I think the smart money is to assume any processor is vulnerable to some variant of this, unless it's definitively explained why not.

Important links from the US-CERT advisory:
Vulnerability Note VU#584653
Microsoft's Advisory
Mozilla's blog post

Side note: I still think Intel's press release was the most weasel-worded, ham-handed attempt at blame deflection I've seen in a long, long time.
 
  • Like
Reactions: skinned66, bowmasters, nt5672 and 3 others
GeminiCricket said:
From AMD's statement:
"AMD is not susceptible to all three variants."
If written by a native English speaker, this is a VERY curious wording. Does this really mean that AMD is not susceptible to ANY of the three variants, or does it mean that AMD might be susceptible to one or two of the variants?
Click to expand...
As a native English speaker I interpret that as "We are not vulnerable to these but could be others". I think it is a native speaker trying to show how perfect AMD and mislead the majority of readers but without outright lying.
 
  • Like
Reactions: O.N.Y.X
timeconsumer said:
Isn't it possible that they could have known but kept quiet on it long enough for this planned sale to go through? I don't follow stocks/trading so I may be reaching, but I'm just curious.
Click to expand...
You are basically speculating that intel employees have been doing things that could mean serious jail time to keep their boss happy, acting significantly against the interests of their company, and that their boss expects that to stay secret?

Employees eventually become ex-employees and talk. One British minister went to jail when his wife became his ex-wife and talked.

The CEO would have to be an absolute imbecile to risk jail time and his career over this, in a way that just must come out.
 
  • Like
Reactions: nwcs and timeconsumer
gnasher729 said:
You are basically speculating that intel employees have been doing things that could mean serious jail time to keep their boss happy, acting significantly against the interests of their company, and that their boss expects that to stay secret?

Employees eventually become ex-employees and talk. One British minister went to jail when his wife became his ex-wife and talked.

The CEO would have to be an absolute imbecile to risk jail time and his career over this, in a way that just must come out.
Click to expand...
According to this article he was aware of the chip vulnerability when he sold off the stock.

http://www.businessinsider.com/inte...fter-company-was-informed-of-chip-flaw-2018-1
 
  • Like
Reactions: jamesrick80
Rob_2811 said:
Not really. My laptop is so thin it probably never gets anywhere near peak performance for the I7 its running anyway. 30% thinner than the previous generation MBP though :cool:
Click to expand...

So true, i realised how awful my MBP2016 15" is when i ran a CryptoNight miner to see it working at max performance.
After a few minutes it slows down and is much slower than my MBP2011. Didn't even bother testing performance in SolidWorks for network rendering after that, just upgraded my modular Tower-PC.
But yeah, it looks really nice and professional with the space grey aluminium.
 
  • Like
Reactions: dantroline
keysofanxiety said:
It was a rule 10b5-1 sale which means it had been planned well in advance and the rule is there to prevent inside trading.

The timing is just very unfortunate.
Click to expand...

I may not be trader (only the wife has taken the Series 7, 63, 9 & 10 so I could ask her for clarification later on I suppose), but there's nothing I can see in § 240.10b5-1 that specified a specific timing of stock sales. Apparently Intel (and others) were told about this last June - ample time to rig up a stock sale. (
https://www.gpo.gov/fdsys/pkg/CFR-2017-title17-vol4/xml/CFR-2017-title17-vol4-sec240-10b5-1.xml)

Westside guy said:
Okay, now that the actual security details are out, finally... it may very well turn out to affect AMD and every other processor maker in existence. The CERT advisory lists AMD as "affected". And given that it's a timing attack, and AMD chips do speculative execution (as does every modern processor) - it's way to early too make the claim it's Intel-only.
Click to expand...


Let's get something straight here; there are TWO separate issues here.

1) Meltdown (CVE-2017-5754 aka Variant 3): This is the easy to exploit flag that affects almost ALL modern day Intel chips - and a few ARM ones. It does NOT - I Repeat NOT affect AMD Chips.
2) Spectre (CVE-2017-5753 aka Variant 1 & CVE-2017-5715 aka Variant 2): This bug is harder to exploit but DOES affect all the major CPU manufacturers in varying degrees (e.g. only a subset of CPU's from these manufacturers are affected).

Meltdown is the one most folk are talking about because of the performance impact on host systems that the fix will implement. It allows applications to read protected kernel memory with apparent ease.

Spectre affects a subsection of CPU's from all the vendors - and, if exploited, allows an attacker to potentially steal data from the kernel hypervisor upwards (hypervisors as well). That said, the exploit is currently much much harder to craft - which is it's only saving grace as of now. However expect that to change in the coming weeks.

Of the two fixes, Meltdown is an easy fix, with a huge performance impact as it cannot be fixed with a microcode patch. Spectre is still a question mark as its fix is going to be harder to craft. Performance impact of Spectre fixes is unknown as of writing, however it may be patchable with a microcode update.

AMD have stated that they believe NONE of their CPU's are susceptible to Variant 2, however Variant 1's still up in the air and they're awaiting more information. Researchers have stated however than the Ryzen chips are affected and Google confirmed that the FX and PRO derivatives are susceptible in certain configurations.

sevvere said:
So we have the same “flaw” in two different chip architectures. ..... I stand firm that this was a three letter agency backdoor.
Click to expand...

Nope, two different flaws. Time to remove that tin-foil hat again.

samcolak said:
All Apple devices (even the new iMac Pro) are affected. Anything with an ARM processor -> https://developer.arm.com/support/security-update

iPhone X A11 Bionic is a derivative version of the Cortex A32. This is as big as it gets - ARM hasn't released info as yet regarding this specific variant as I suspect it might have significant impact on Apple.
Click to expand...

Please stop being so bloody hysterical! :rolleyes: The A32 is NOT - I Repeat NOT on the list of CPU's that ARM have identified as being affected by Spectre: The FULL list of ARM CPU's susceptible to Spectre is: Cortex-R7, Cortex-R8 [these two CPU's are highly specialized CPU's and not in any standard gear], Cortex-A8, Cortex-A9, Cortex-A15, Cortex-A17, Cortex-A57, Cortex-A72, Cortex-A73, and Cortex-A75 [as used in Qualcomm's new Snapdragon 845].

In addition, ARM identified the A15, A57 and A72's are ALSO being susceptible to Meltdown.

Until ARM have identified the A32 (or indeed any of the Apple derivatives) you're just guessing based on ignorance of the problems identified. o_O
 
Last edited:
  • Like
Reactions: MrAverigeUser, Val-kyrie, Mescagnus and 4 others
No mention of IBM or Motorola... does that mean my Power Mac G5 file server with it's IBM 970MP is impervious to this exploit? I would reckon probably so... after all the whole world even Apple itself forgot that we even existed.
 
Intel has announced today their hack proof, brand new 9th Gen Core iAnalog processor....

1949-electronic-analog-computer.jpg
 
  • Like
Reactions: Bacillus
You must log in or register to reply here.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.
Top Bottom
Back