Intel Chips Have Memory Access Design Flaw and Fix Could Lead to Performance Drop

It seems that in order for this processor bug to be exploited, one would have to have run a malicious application or have a machines setup with other vulnerabilities which would allow remote access to run a malicious program to exploit the bug. That said, for someone like me who doesn't grant remote access to other people to their machines, and who runs only a few programs from reputable developers, e.g., Adobe, Bare Bones Software, Bombich, etc., there seems to be no chance my security will be compromised. That being the case, I hope if there is a security software update that does end up slowing machines by 5 to 30% as some have said it will, that this security feature can be turned off for machines like mine where the security risk is so minimal or really, nonexistent. I'd much rather keep my machines faster rather than be protected from a threat I will never have to face.

Just curious if my understanding of all this is correct, or am I missing something?

 

I wonder if we see a MacBook with an A11X this year (WWDC)?

The A11 is already smoking fast, and an A11X with improved GPU and better thermals could easily keep up with the Intel processors Apple currently uses for its lower-end models.

Imagine a device with 20 hour battery life and standby time measured in weeks.

 

Poor Intel. It's a bad day for them.

 

It would be great and I think we'll get to that in a few years, but Intel still plays an important part on the Mac, they could release the 12'' Macbook and a new MBA with their SoC but I'm not sure they're ready for the MBP and for the iMac, so at least for some years we'd have some machines with Intel some others with ARM

 

Interesting tweet:
"The question on everyone's minds: Does MacOS fix the Intel #KPTI Issue? Why yes, yes it does. Say hello to the "Double Map" since 10.13.2 -- and with some surprises in 10.13.3 (under Developer NDA so can't talk/show you). cc @i0n1c @s1guza @patrickwardle"

https://twitter.com/aionescu/status/948609809540046849

 

Other sites are reporting that MS has been developing and testing a patch with testers since November.

 

Class action lawsit(s) in 3,2,1....

 

Up to 30% performance drop?!

 

Everybody thinks this is bad news for Intel, but actually it all depends on how quickly they can fix the problem on new chips entering production. If they can turn around it quickly, they probably stand to benefit the most.
Any idea of them being sued by anyone is pure nonsense, unless it can be proved that there was malicious intention on their behalf. I have been writing and selling software and IT services all my life and we all know that everything has bugs somewhere, it is just a case of finding them.

 

intel also abolished moore's law

x86 CPU progress has slowed to a crawl

 

That depends entirely on your point of view of this issue.

 

If so, that's going to be a very long list....

 

• The Volkswagen case was deliberate. This is not.
• The Volkswagen case had nothing to do with horse power, so it's not clear why you'd bring that up.
• Even if it did, the equivalent would be for an Intel's specs — e.g. MHz — to drop. Rather, the results (which aren't part of the spec) do.

 

It's been speculated that JavaScript on a webpage could be a possible vector for this.

 

"Finally, macOS has been patched to counter the chip design blunder since version 10.3.2, according to operating system kernel expert Alex Ionescu. And it appears 64-bit ARM Linux kernels will also get a set of KAISER patches, completely splitting the kernel and user spaces, to block attempts to defeat KASLR. We'll be following up this week."

Is it a typo then; that macOS has been fixed since 10.3.2? Anyway, I'm running 10.13.2, and I haven't noticed it being horrifically slow, so I'm guessing all's well.

[EDIT: Whoops! Duh... yes, the linked article mentions 10.13.2... BTW, I didn't think that macOS somehow magically knew about it all those years ago, but just wondered if maybe the kernel design they used was somehow different.]

 

According to The Register, macOS as been patched...

https://www.theregister.co.uk/2018/01/02/intel_cpu_design_flaw/

"Finally, macOS has been patched to counter the chip design blunder since version 10.3.2, according to operating system kernel expert Alex Ionescu. And it appears 64-bit ARM Linux kernels will also get a set of KAISER patches, completely splitting the kernel and user spaces, to block attempts to defeat KASLR. We'll be following up this week."

--- Post Merged, Jan 3, 2018 ---

From Alex Ionescu Twitter account...

https://twitter.com/aionescu/status/948609809540046849

 

Introducing the 2018 Mac Pro:

 

Liking it! A Mac Pro tower wouldn't that be something!

Its not that simple to port OSX to an AXX CPU. It might share code with iOS but its not as simple as plonking the new chip in. It took years remember when SJ announced the shift to intel cpus.

RE intel

Intel: a confusing product line, mediocre improvement year on year, dropping tick tock, forced socket changes when arch barely changed...and so on.... Is the turn of the tide here aka Pentium 4 V AMD Athlon? Lets see.

 

That is absolutely shocking. A performance drop? WHAT?

 

They should go BACK to RISC... Power 9 anyone? If they did, the performance would be stellar at the least!!!
Who knows, maybe Apple has something else in store here????
Please Apple... go RISC!!!!!

 

NO... RISC!

 

First, It's all very early. As things unfold, don't forget that.

I did not notice this news as it broke last night. I was deep in trying to get some Windows Servers and Windows 10 to play nice to each other. I was not able to do it. The reason I could not, is interesting today with this news. Microsoft is starting to prevent you from using Windows on systems with older processors, not because they want you to need to buy a new system and then they could sell you a Surface, no they are adding features to their OSes that need certain processor features to be able to work. What happens if you don't have a new enough processor. Simple, you can't install the new version of Windows. We are seeing this first on Windows Server but they have said it's coming to Windows 10 as well. It looks like they are going to do this in both directions. New systems will not run old versions of windows https://www.theverge.com/2016/1/16/10780876/microsoft-windows-support-policy-new-processors-skylake and as we are seeing in Windows 2016 Server, it will not install if you don't have the correct processor features. (In this case SLAT). So Microsoft is looking to insure that you are using the newest software (Duh) and also the latest hardware. I am sure Intel is happy about that. With this news I am sure that Intel will be pushing their major Software partners to push this idea (obsoleting your system) even harder. Intel wants these CPU's to go away. If they could make Microsoft or Apple the bad guys that would be great. OK, so this is MAC Rumors so where does this all leave Apple.

First, might this cause them to look harder at the idea that Mac's should move to ARM. Yes it will. But I don't think that will happen. Not totally. There might be a ARM Mac someday. But I do not see Apple giving up on the X86 platform for some time. Why? lots of reasons. First you would have to BUY all new software. Developers would have to write all new software. The rumor about a common app platform across ios and Mac OS would help, but customers still want to run the software they have. And don't say emulation. We don't want emulation. By the way, know what the most popular application is on Macs. WINDOWS. So X86 is not going anywhere. Not now. Microsoft is doing work to run Windows on ARM , so maybe someday, ARM will make more sense for all of us.

Also, let's talk legal. There will be lawsuits. There are ALWAYS lawsuits. But what you can file and what you can win are not the same thing. I am not yet sure that there is much to sue about here. We don't yet know enough about this. But I own several computers with these CPU's and I do not think I will ever see much of anything out of the legal avenue.

Yes this is a big deal. We do not yet know how big of a deal this is. We don't know the full problem and how you fix it. We don't know what that fix will cause. We don't know how OS makers and Computer makers are going to respond. We don't know how Intel is going to respond.

That all takes me back to where I started. It's early. Someone earlier in the thread suggested popping some popcorn and sitting back and watching. That sounds like the best advice at this point.

 

Apple has already made multiple architecture transitions, despite customers having to buy new software and developers having to run new software. That won’t stop them. Mac started on 68k, then went to powerppc, then went to Intel x86, and now has transitioned to AMD64 (ie x86-64). Each of these transitions have caused the problems you refer to, but Apple did each of them because they were the right thing to do at the time.

 

Smaller market share, now pretty much every Mac in use is intel based across not just a select computer user base [IE remember that the moves mentioned where when Apple was really catering for a small amount of personal users and many professional users]...

However, if apple is trying to merge app stores..either some Xcode tool to convert or a possible arch change...

 

Intel has a tool on their website to check if your CPU is vulnerable but there is no Mac OS version!