Monday, January 22nd 2018

Intel Announces Root Cause of Meltdown, Spectre Patch Reboot Issue Identified

Intel has finally come around towards reporting on the state of the reboot issues that have been plaguing Intel systems ever since the company started rolling out patches to customers. These patches, which aimed to mitigate security vulnerabilities present in Intel's chips, ended up causing a whole slew of other problems for Intel CPU deployment managers. As a result of Intel's investigation, the company has ascertained that there were, in fact, problems with the patch implementation, and is now changing its guidelines: where before users were encouraged to apply any issued updates as soon as possible, the company now states that "OEMs, cloud service providers, system manufacturers, software vendors and end users stop deployment of current versions, as they may introduce higher than expected reboots and other unpredictable system behavior." A full transcription of the Intel press release follows.
"As we start the week, I want to provide an update on the reboot issues we reported Jan. 11. We have now identified the root cause for Broadwell and Haswell platforms, and made good progress in developing a solution to address it. Over the weekend, we began rolling out an early version of the updated solution to industry partners for testing, and we will make a final release available once that testing has been completed.

Based on this, we are updating our guidance for customers and partners:
  • We recommend that OEMs, cloud service providers, system manufacturers, software vendors and end users stop deployment of current versions, as they may introduce higher than expected reboots and other unpredictable system behavior. For the full list of platforms, see the Intel.com Security Center site.
  • We ask that our industry partners focus efforts on testing early versions of the updated solution so we can accelerate its release. We expect to share more details on timing later this week.
  • We continue to urge all customers to vigilantly maintain security best practice and for consumers to keep systems up-to-date.
I apologize for any disruption this change in guidance may cause. The security of our products is critical for Intel, our customers and partners, and for me, personally. I assure you we are working around the clock to ensure we are addressing these issues.

I will keep you updated as we learn more and thank you for your patience." Sources: Intel NewsRoom Reboot Issues, Intel newsRoom Udpated Guidance
Add your own comment

41 Comments on Intel Announces Root Cause of Meltdown, Spectre Patch Reboot Issue Identified

#1
rtwjunkie
PC Gaming Enthusiast
jsfitz54 said:
Question: Can the Patch be undone, for those that had it auto applied? (((OR))) Will it stay in place for some reason and leave behind remnants?
I think technically it can, because for me it repeatedly failed at the 85% or so point, then went through a big production of rolling me back.
Posted on Reply
#2
kn00tcn
Aquinus said:
Yet, this is update #2 this month.

Undoing the patch for meltdown because it can crash certain systems? I'm not sure how that qualifies as "not so bad." :laugh:
...update #2 of the same major topic, when have there been multiple quick microcode updates PREVIOUSLY that didnt align to product launches?

crashes are binary, so bad implies there is some amount of bad that can be accepted, but this is 'some reports came in so ctrl+z', ubuntu recently had some laptop firmware losing issue that no other distro had, so sometimes it's not even the microcode's problem but what the distro was trying to do (edit: was update #1 this or was it still meltdown?)

OSdevr said:
If Ubuntu was a server distro I might understand pulling the update for 'mission critical' reliability reasons but it's a desktop distro, and as the most popular Linux distro (and one which I personally use) others are likely to follow suit. There are a lot of distros based on Ubuntu, it's not some no name.
ubuntu dropped their unity work & switched to gnome made by red hat, already on systemd made by red hat, people dont 'follow' ubuntu, they follow upstream changes that may or may not match what ubuntu does (if it isnt obvious, i dont like ubuntu or the way some packages are handled)

Katanai said:
As you can see, I have been called a moron, a fool, people like me are dangerous etc. I refuse to be humble in front of such people, who should be eating their words right now. I can tell you to rest assured that my Operating System is fully under my control, my browser is not losing any performance
who cares what a few people said on some forum? at a subconscious level it's boasting about your setup so it adds too much filler text (yes, ironically i want an evolved reaction from you even though we have nothing to do with each other & nobody cares tomorrow... if anything, not feeling the need to be defensive will bring you peace while the others stay stressed, letting go of an argument is applicable to both sides not just the attacker (disclosure: i saw your old posts the first time though it's good that i had a few days to think about them more & reread them today))

your browser is losing performance because they loosened a certain timer latency, not that it will be noticable, neither is the touted 30% loss that people keep reposting... that loss is extremely specific & probably most obvious on an SSD where speeds are crazy high, 0.1s to 0.5sec is a multi hundred percent loss, but still less than a second, so i dont avoid updates without measuring or at least noticing losses for myself (even if they exist, in a game example, 90 to 85 fps is irrelevant, winxp might be faster sometimes, but not enough to matter, deferring updates could be in the same situation)

edit: by the way, work has been done since the summer, but i guess a lot of rushing was going on in january
Posted on Reply
#3
Katanai
R-T-B said:
For my part, I didn't call you a fool. I said your actions were foolish. There is a very big difference.
Let me teach you something today. This is the definition of foolish: "lacking good sense or judgement; unwise."

Now please tell me which one of us was the one acting foolish. Me, using my judgement to determine that it would be wiser to wait until this is sorted out? Or you, downloading and deploying untested beta software designed to gimp CPU's, that now even the ones who made it say it would be better to uninstall? It's one thing to call someones actions foolish and then falsely pretend that the actions of someone doesn't say anything about his character and somehow his actions are separated from him and you can belittle them at your own leisure without insulting him by one bit. It's something else though when your own actions fall under the exact definition of the words you projected on someone else while his are the exact opposite. This says a lot about the character of a person who goes out and projects his own flaws unto others. Please, in the future, instead of worrying about what other people do and judging their actions, take some time out and think better about yours. Use the time you would waste trying to judge others to improve yourself, because it seems that you really need it. This advice goes out to [USER=65960]kn00tcn[/USER] too because he, as any good Canadian does, seems to be more interested in other's people's business than his own...
Posted on Reply
#4
R-T-B
Katanai said:
It's one thing to call someones actions foolish and then falsely pretend that the actions of someone doesn't say anything about his character and somehow his actions are separated from him and you can belittle them at your own leisure without insulting him by one bit.
Everyone makes foolish decisions at some point. This does not make them fools. The definition actually says:

(of a person or action) lacking good sense or judgment; unwise.

Note the bold.

Look, if you want to take offense, I can't stop you. I can only state my intent and hope for the best.

The rest of your post is full of unnecesary implications that I feel would be counterproductive to reply to.
Posted on Reply
#5
Katanai
R-T-B said:
Everyone makes foolish decisions at some point. This does not make them fools.
Here you are again projecting your own flaws unto others. This time on the whole humanity. You have made a mistake and were called out for it. What have you learned from this? Maybe try to better yourself? No! Everyone makes mistakes and there's nothing wrong with that. That's what you are telling me right now and that's why I think you will never learn.

R-T-B said:

Look, if you want to take offense, I can't stop you. I can only state my intent and hope for the best.
You go on by refusing to take any responsibility for your own actions. You said something that someone else found offensive. Of course it's not your fault, other people are taking offense and running away with it and you can't stop them. Let me tell you something, if you spill coffee on someone and he jumps, although you didn't mean to do that, you say you are sorry. If you say something and someone is offended by it, although that was not your intention, you don't have to say sorry but at least you say something like: I didn't mean to offend. There, I've showed you how you can stop people from taking offense. Because the problem here is that you don't even "state your intent" like you say. When you took sides with someone that called me a moron and you called my choice of action foolish what was your intention? To amuse me? To save the planet?

R-T-B said:

The rest of your post is full of unnecesary implications that I feel would be counterproductive to reply to.
The rest of my post, like this one right here, is filled with truth. I feel pity for anyone that describes truth as "unnecessary implications" and refuses to confront it. I repeat my advice to you: spend some time thinking about your behavior and how your actions impact and are perceived by those around you. Try to confront your own problems and resolve them yourself instead of projecting them unto others. As long as you go around making mistakes and then thinking: well everybody does that, you will not be able to truly advance...
Posted on Reply
#6
R-T-B
This whole discussion frankly says a lot more about you than me.

You seem a very angry person. For that reason, I'm out.
Posted on Reply
#7
Katanai
R-T-B said:
This whole discussion frankly says a lot more about you than me.

You seem a very angry person. For that reason, I'm out.
Of course yet again it's someone else's fault. Although nothing in my discourse was in any way or form uncivilized and could have been perceived as angry, once again you project something from inside yourself on someone else. Maybe what I've said has made you angry so in your head I'm the angry one now. Because that's what you do and sadly it seems will continue to do into the future. It seems there's nothing I can do to change that so: Bye! Enjoy your gimped CPU! :D
Posted on Reply
#8
R-T-B
Katanai said:
Of course yet again it's someone else's fault.
No, I don't think it's anyones fault. I think it's just a sucky situation I don't want to be a part of.

I'm sorry I made you mad. It wasn't my intention but it is what happened. Truth be told, I don't feel too guilty about it as it seems you are abusing definitions to make yourself angry and I never really intended insult, so what more can I do but say "sorry, didn't mean to make you mad bro?" I can't help but infer that that may mean you are an angry person, if you aren't and are thinking of rainbows right now, please be pleased to know I am happy for you.

Good day.
Posted on Reply
#10
Katanai
R-T-B said:
No, I don't think it's anyones fault. I think it's just a sucky situation I don't want to be a part of.

I'm sorry I made you mad. It wasn't my intention but it is what happened. Truth be told, I don't feel too guilty about it as it seems you are abusing definitions to make yourself angry and I never really intended insult, so what more can I do but say "sorry, didn't mean to make you mad bro?" I can't help but infer that that may mean you are an angry person, if you aren't and are thinking of rainbows right now, please be pleased to know I am happy for you.

Good day.
When you say "I'm out" and then you come back again that only means one thing: the other person is making you angry. It also means that you are not a man of your word but that's another story. Your every action in this thread proves my words are true and everyone can see it. This article itself under which we are commenting here proves that I was right from the start and you were utterly wrong. I have no reason to be angry, quite the contrary, you have all the reasons in the world. If it wasn't clear from the smiley I posted in my message, I am actually smiling while typing this. :laugh: I'm having an argument with someone on a forum called techpowerup that instead of trying to get more performance out of his CPU used software to gimp it and advised me to do so too so we can save the planet together. Now the performance charts come out so he can see how much he gimped it and he likes it. :laugh: Here, just so you can see I'm not mad at you, I will try to teach you one more thing. This is how it's done son: School session is over. Peace, I'm out!
Posted on Reply
#11
R-T-B
I came back because you were in my notification hotbar, and thought maybe the words "I'm sorry" should be uttered whether I saw the damage or not.

So I'm sorry. I did not mean to hurt your feelings by telling you a security patch may be important.

I'm glad you are happy. Enjoy your 5-10% performance edge and be careful with that security hole.

By the way, this microcode patch was not deployed to Ryzen CPUs, which I am using. So... yeah. I've really got no point except to offer my advice and feelings on best practices. Yes, I feel you are being foolish still. I am sorry that offends you but my thoughts are my own.
Posted on Reply
#12
RCoon
Gaming Moderator
Remain on topic or reply bans will be issued. Use the ignore functionality, it's proper good like.
Posted on Reply
#13
Aquinus
Resident Wat-man
OSdevr said:
Ubuntu Server is. It is distinct from Ubuntu Desktop.
Um, what? Maybe you can explain to me exactly what those differences are because Ubuntu Desktop 17.10 and Ubuntu Server 17.10 use the same software packages, the same kernels, the same drivers, etc. Some of the kernels used to be distinguished but, even that isn't the case anymore. Example:
My tower (running Ubuntu "Desktop" 17.10):
code:
$ uname -a
Linux kratos 4.13.0-31-generic #34-Ubuntu SMP Fri Jan 19 16:34:46 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux

...and the tower in the attic crunching with Ubuntu "Server" 17.10:
code:
$ uname -a
Linux smite 4.13.0-16-generic #19-Ubuntu SMP Wed Oct 11 18:35:14 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux

So, other than the fact that I haven't updated Smite in a while, they're using the same kernel and the same packages, yet one is "desktop" and the other is "server". The only difference is what packages are installed when the OS was installed for the first time but, I can install Ubuntu Server, run "sudo apt install ubuntu-desktop" and there is literally no difference from Ubuntu Desktop. In fact, every package is addative, it doesn't remove packages when going to desktop, so what gets installed on server is just a subset of what gets installed on desktop. Likewise, if I did a desktop installation and removed just about every package for anything GUI-related and that Ubuntu Desktop will be in [roughly] a state that is the same as a Ubuntu Server installation (uninstalling packages is never as clean as never installing them.)

Either way, Ubuntu Server is not distinct from Ubuntu Desktop, the only difference is what you start with. Even the one thing that used to be different, isn't anymore (the kernel.) Even the /etc/apt/sources.list are practically identical for all Ubuntu apt repos. So, I'm calling foul. There is a reason why the alternative installer doesn't exist anymore, because that's the GUI-less server installer. It's really not any different. In fact, if you want a minimal or mimimal VM installation, desktop or not, you use the server installer but, there is legit no difference between the end result other than what packages are there after installation but, I can assure you that every package in a base server installation is the same as desktop when you talk about Ubuntu's apt repos.
Posted on Reply
#14
OSdevr
Aquinus said:
Um, what? Maybe you can explain to me exactly what those differences are because Ubuntu Desktop 17.10 and Ubuntu Server 17.10 use the same software packages, the same kernels, the same drivers, etc. Some of the kernels used to be distinguished but, even that isn't the case anymore. Example:
My tower (running Ubuntu "Desktop" 17.10):
code:
$ uname -a
Linux kratos 4.13.0-31-generic #34-Ubuntu SMP Fri Jan 19 16:34:46 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux

...and the tower in the attic crunching with Ubuntu "Server" 17.10:
code:
$ uname -a
Linux smite 4.13.0-16-generic #19-Ubuntu SMP Wed Oct 11 18:35:14 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux

So, other than the fact that I haven't updated Smite in a while, they're using the same kernel and the same packages, yet one is "desktop" and the other is "server". The only difference is what packages are installed when the OS was installed for the first time but, I can install Ubuntu Server, run "sudo apt install ubuntu-desktop" and there is literally no difference from Ubuntu Desktop. In fact, every package is addative, it doesn't remove packages when going to desktop, so what gets installed on server is just a subset of what gets installed on desktop. Likewise, if I did a desktop installation and removed just about every package for anything GUI-related and that Ubuntu Desktop will be in [roughly] a state that is the same as a Ubuntu Server installation (uninstalling packages is never as clean as never installing them.)

Either way, Ubuntu Server is not distinct from Ubuntu Desktop, the only difference is what you start with. Even the one thing that used to be different, isn't anymore (the kernel.) Even the /etc/apt/sources.list are practically identical for all Ubuntu apt repos. So, I'm calling foul. There is a reason why the alternative installer doesn't exist anymore, because that's the GUI-less server installer. It's really not any different. In fact, if you want a minimal or mimimal VM installation, desktop or not, you use the server installer but, there is legit no difference between the end result other than what packages are there after installation but, I can assure you that every package in a base server installation is the same as desktop when you talk about Ubuntu's apt repos.
It appears you are correct. My mistake.
Posted on Reply
#15
ExV6k
I might be just a little bit off-topic, but what about Sandy Bridge? Did intel update those chips and if so, is it safe to assume they're not affected by this random rebooting issue?
Also, I've read somewhere that the first microcode update caused some chips (specifically Haswell and Broadwell?) to report "internal errors" on HWInfo64, can anyone with an updated microcode confirm this, please?
Posted on Reply
#16
Arctucas
FordGT90Concept said:
I've applied the microcode update to my 6700K mobo and still have not encountered any issues.
Neither have I, nor any significant slowdown.

I was simply wondering if there were different microcodes for different processors that perhaps had different effects?
Posted on Reply
Add your own comment