Monday, June 26th 2017

Critical Flaw in HyperThreading Discovered in "Skylake" and "Kaby Lake" CPUs

A critical flaw was discovered in the way Intel implemented its simultaneous multi-threading technology, HyperThreading, on "Skylake" and "Kaby Lake" processors. Being a micro-architecture specific flaw, this could affect all implementations, from low-power mobile chips, to mainstream desktop, high-end desktop, and perhaps even enterprise-segment Xeon processors. At this time, there are no security implications of this flaw.

Intel chronicled this flaw in its micro-architecture errata "SKZ7/SKW144/SKL150/SKX150/SKZ7/KBL095/KBW095," and described it as follows: "Under complex micro-architectural conditions, short loops of less than 64 instructions that use AH, BH, CH or DH registers as well as their corresponding wider register (e.g. RAX, EAX or AX for AH) may cause unpredictable system behavior. This can only happen when both logical processors on the same physical processor are active." As an implication, Intel goes on to note that Due to this erratum, the system may experience unpredictable system behavior."
The HyperThreading flaw can be fixed through a micro-code update distributed as a UEFI firmware update. Typically, it becomes the responsibility of DIY PC motherboard, pre-built desktop, and notebook manufacturers, to distribute the update. The issue first came to light in a Debian Linux user mailing-list, although it affects all PC operating systems, not just Linux. Support groups of Debian recommend disabling HyperThreading in the UEFI setup programs of your computers as a temporary workaround, till the micro-code patch is applied. Disabling HyperThreading will reduce performance in multi-threaded apps.
Source: Debian Mailing List
Add your own comment

98 Comments on Critical Flaw in HyperThreading Discovered in "Skylake" and "Kaby Lake" CPUs

#1
RejZoR
Intel is really getting a shitstorm shoved straight in their face. I bet AMD is chuckling evilishly in the corner :D
Posted on Reply
#3
ssdpro
Do we have any cases of anything negative happening related to this? I can't believe we should really disable functionality based on nameless "Debian support groups" rather than Intel's specification update errata. Skylake has been around a long time now and forums show remarkable stability with CPU related tasks.
Posted on Reply
#4
P4-630
Glad I'm fine then with my i5 with no hyper threading.:)
Posted on Reply
#5
trparky
Microsoft will probably issue a patch to the OS itself that loads new microcode at load time. It's not the first time Microsoft has done such a thing. It's happened before (example).
Posted on Reply
#6
Frick
Fishfaced Nincompoop
ssdproDo we have any cases of anything negative happening related to this? I can't believe we should really disable functionality based on nameless "Debian support groups" rather than Intel's specification update errata. Skylake has been around a long time now and forums show remarkable stability with CPU related tasks.
caml.inria.fr/mantis/view.php?id=7452

I don't pretend to understand it, but well.
Posted on Reply
#7
Solaris17
Super Dainty Moderator
ssdproDo we have any cases of anything negative happening related to this? I can't believe we should really disable functionality based on nameless "Debian support groups" rather than Intel's specification update errata. Skylake has been around a long time now and forums show remarkable stability with CPU related tasks.
then lets look where the well trusted debian group got it.

www.intel.com/content/dam/www/public/us/en/documents/specification-updates/7th-gen-core-family-spec-update.pdf
Posted on Reply
#8
CAPSLOCKSTUCK
Spaced Out Lunar Tick
No problems here


Long live socket 1366..........:peace:
Posted on Reply
#9
meirb111
why now?
skylake and kaby lake are out for a long time ,so how was it not found before?
Posted on Reply
#10
EarthDog
Its so critical, yet, what is actually wrong? The article here doesn't say much at all, sadly. Thank you, Frick, for taking the time to do that, for free...
trparkyMicrosoft will probably issue a patch to the OS itself that loads new microcode at load time. It's not the first time Microsoft has done such a thing. It's happened before (example).
Or, as was said, they will do it in a microcode update in the form of a new bios from AIBs/OEMs...
Posted on Reply
#11
trparky
Enthusiast motherboards may get the BIOS/UEFI update to fix it but as for the OEMs like HP, Dell, etc.? Fat chance!
Posted on Reply
#12
EarthDog
They offer them as well... it will depend. Typical delivery however is through UEFI... there are of course exceptions. :)
Posted on Reply
#13
trparky
Apparently you have more faith in the OEMs than I do, which is none.
Posted on Reply
#14
EarthDog
Heh, yeah, I take many people's opinions with a grain of salt... some, a salt block. :)
Posted on Reply
#15
trparky
Most OEM hardware is less than shit. There's a reason why I built my own PCs.
Posted on Reply
#16
xkm1948
This should affect the new Core I9 X processors as well. Oops.
Posted on Reply
#17
Frick
Fishfaced Nincompoop
trparkyMost OEM hardware is less than shit.
Well, no. Not really.
Posted on Reply
#18
Tomgang
Just another good reason to stay on X58 for a while longer. X58 never failed or been a pain in the ass and its not like X58 lagging features i need.

Intel really messing up one thing after another lately. Some Atom cpu had a flaw, skylake X whas a rushed messed up lauch and now hyper trading is being a pain in the ass, intel else have used HT since the Pentium 4 days so is´t nothing new for them. But thats what you get when you become lazy, greedy and just pissing on customers.

Just waiting for my GTX 1080 TI to come and my X58 system upgrade is now complete for a year or two, then i am properly going on a new system.

Until then you can suck it intel. I am not gonna pay lazy bums then they just fucking things up any way and rips your wallet.
Posted on Reply
#19
Steevo
AMD "We are making a new Ryzen spin to fix minor bugs, also have this BIOS to push performance up to help close that 5% gap at half the price and noob level overclocking with power savings too"

HOLY CRAP THEY ARE SCREWING US ALL HOW ABOUT US BETA TESTERS FOR AMD I CAN'T BELIEVE THEY WOULD DO THIS!!!

Intel "HT is broken for some things, also here is the same processor with MORE CORES and same issues higher power consumption and more thermal limits and no overclocking supported, also make sure you match your CPU to the configuration you want for PCIe lanes and memory or else"

Its OK, they will patch it with BIOS updates, everyone should be OK with BIOS updates and known issues...


Amazing.
Posted on Reply
#20
Shihab
TomgangJust another good reason to stay on X58 for a while longer. X58 never failed or been a pain in the ass and its not like X58 lagging features i need.

Intel really messing up one thing after another lately. Some Atom cpu had a flaw, skylake X whas a rushed messed up lauch and now hyper trading is being a pain in the ass, intel else have used HT since the Pentium 4 days so is´t nothing new for them. But thats what you get when you become lazy, greedy and just pissing on customers.

Just waiting for my GTX 1080 TI to come and my X58 system upgrade is now complete for a year or two, then i am properly going on a new system.

Until then you can suck it intel. I am not gonna pay lazy bums then they just fucking things up any way and rips your wallet.
At the risk of sounding an Intel apologist, bugs happen, we can't really fault intel much here. Sure, someone was sloppy with their work, but it doesn't really have that much an effect to drive this negative a reaction. If intel bugs has been considerably frequent, it would've been a different story, but it isn't. Moreover, according to Frick's link, the fix for this issue is already out there.
Posted on Reply
#21
oxidized
RejZoRIntel is really getting a shitstorm shoved straight in their face. I bet AMD is chuckling evilishly in the corner :D
And you're happy, are you not? :):):)
Posted on Reply
#22
Filiprino
It is called Debian GNU/Linux, not Debian Linux. It is not a Linux made by Debian, but a GNU/Linux operating system put together by Debian.
Source.

A large part of the basic tools that fill out the operating system come from the GNU project; hence the names: GNU/Linux, GNU/kFreeBSD, and GNU/Hurd. These tools are also free.


Intel has been shoddy here. Not talking about this bug nor contacting with those who reported it to them. The microcode was released silently.
Posted on Reply
#23
Solaris17
Super Dainty Moderator
FiliprinoIntel has been shoddy here. Not talking about this bug nor contacting with those who discovered it. The microcode was released silently.
lol?

Intel is more transparent with errata than AMD ever has. Its been there motto since forever! You can download the PDF errata documents of any intel CPU product family right from there website?

Silly consumers. This happens all the time. it just doesnt normally affect home users because the software is either IE resilient or it doesnt trigger the issue. Notice how the people that found it were running server teams with large data sets?

Does the forum consensus on TPU actually think CPUs are made without fault other than missing or wanted feature sets by consumers? every single CPU revision has a published errata.

This isnt shady at all. They do it with everything, if you can get your hands on/if the company publishes the documents. from chipsets to hard drive micro-controllers.

EDIT:: Lets take a look

support.amd.com/TechDocs/51810_16h_00h-0Fh_Rev_Guide.pdf

Looks like AMD does it too, and no plan to fix at that. Where is your god now AMD fans?

Here ya go. I took about 35seconds to shatter your delusions of purity.

support.amd.com/en-us/search/tech-docs#k=errata
Posted on Reply
#24
EarthDog
FrickWell, no. Not really.
I find it incredibly difficult to sway opinion out of fact on this one... same with Hugh Mungus... ;)
Posted on Reply
#25
TheGuruStud
FrickHere's the actual source.

lists.debian.org/debian-devel/2017/06/msg00308.html

Of note is probably how it affects two generations, mostly because that tells you how much of an edge case it is.
That didn't stop Intel paid publications and fanboys from railing AMD over its TLB bug all those years ago (which not a single person ever experienced). Full steam ahead. Choke on it, Intel.
Posted on Reply
Add your own comment
Dec 24th, 2024 14:42 EST change timezone

New Forum Posts

Popular Reviews

Controversial News Posts