Live data from Hacker News

Intel CEO: Patches will come to 90% of chips in the next week

techcrunch.com

91–100 of 137 posts

Re: Intel CEO: Patches will come to 90% of chips in the next week

#91
post #62

Earlier quoted context omitted.

Let me break down what the article really says (it's pretty short): > He also said that Intel expects to issue updates > to its processors soon. More than 90 percent will be > getting them within the week, and the rest by the end > of January. Intel expects 90 percent will "get an update." 90 percent of what? It should be self-evident that Intel means 90 percent in whichever way gives them the largest percentage. I e…

...or maybe the other 10% are in-order non-speculative designs which are intrinsically immune; they still have a few: https://en.wikipedia.org/wiki/Intel_Quark (basically a die-shrink of a 486!) https://en.wikipedia.org/wiki/Bonnell_(microarchitecture)

"Intel expects to issue updates to its processors soon.

... the rest by the end of January."

Not "the rest are not affected," or "the rest are intrinsically immune." If they were, Intel could have deflated a lot of the PR around AMD being immune while Intel is affected.

Re: Intel CEO: Patches will come to 90% of chips in the next week

#92

Earlier quoted context omitted.

Intel's not patching anything. They're relying on Windows, Linux and macOS patches to work around the vulnerability. Presumably that's how the 90% claim can be made. Very disingenuous of Intel tho. EDIT: There are microcode updates included in the OS updates: https://access.redhat.com/articles/3311301

> Very disingenuous of Intel tho This crisis has taken Intel, in my mind, from an American behemoth at the vanguard of technology to a sclerotic overgrown mess. Bugs happen, crises happen. When you're a $200 billion company, those mistakes scale deafeningly. The bugs are unfortunate, but not unreasonable. Intel's communication, however, from the first press release to crap like this, has been disingenuous to the poin…

They even tried to name the macro in Linux something like "INSECURE_KERNEL" instead of accurately describing that the insecurity lies in the CPU, not kernel.

Re: Intel CEO: Patches will come to 90% of chips in the next week

#93
post #65

Earlier quoted context omitted.

> FreeBSD Security Team was notified of the issue in late December Anyone else thinks this was kind of a slap in the face to the smaller communities and companies or is it just me? They were notified in late December, right before the holidays, so that's basically only 2-3 weeks of work. Obviously nobody _had_ to notify anyone, could have just released it right away, so it was a professional courtesy, but why not ext…

The more people you tell, the higher the chance of a leak. "Loose lips sink ships."

FreeBSD isn't two people in a garage, it's a foundational part of the internet developed by professionals who can be counted upon to do the right thing.

Intel and Google don't have a leg to stand on: I'm super glad they found the bugs, but their disclosure has been nothing but a shitshow.

Re: Intel CEO: Patches will come to 90% of chips in the next week

#94
post #28

What I'm worried about is that it will be hard to avoid these security patches when you don't need them. Say you have a non-virtualized, non-shared server that only runs your own trusted code. I don't want to be forced to pay the performance penalty but it might be unavoidable without resorting to maintaining your own linux fork.

(on linux) there's a way to disable it. AFAIK you add 'kpti=off' to the kernel command-line which is possible, e.g., by pressing 'c', from grub. IDK about other OSes.

Re: Intel CEO: Patches will come to 90% of chips in the next week

#95
post #64
post #45

Earlier quoted context omitted.

I'm giving it one to two generations of CPUs before I do a new build.

But since the OS fixes to protect against meltdown come with a potential performance penalty, AMD is a better choice now than it was a week ago.

Been planning on buying EPYC for our upcoming OpenShift deployment already - now I’ve got a smoking gun to get management onboard.

Re: Intel CEO: Patches will come to 90% of chips in the next week

#96
post #35

According to last weeks press release, they mean 90% of recent chips, which is far less than 90% of intel CPU's out in the wild: https://newsroom.intel.com/news-releases/intel-issues-update... > Intel has already issued updates for the majority of processor products introduced within the past five years. By the end of next week, Intel expects to have issued updates for more than 90 percent of processor products intro…

Presumably the launch date, Intel lists launch dates for all of their processors on ark.intel.com [1]

5 years is Haswell, so Haswell and newer should get the microcode fix. A 3 year old computer is likely to have a CPU no more than 4 years old, but you can look it up to verify. (Both Windows and Linux can show the CPU model, and Intel's Ark site gives the release date.)

[1] https://ark.intel.com/products/series/75023/4th-Generation-I...

Re: Intel CEO: Patches will come to 90% of chips in the next week

#97
post #35

According to last weeks press release, they mean 90% of recent chips, which is far less than 90% of intel CPU's out in the wild: https://newsroom.intel.com/news-releases/intel-issues-update... > Intel has already issued updates for the majority of processor products introduced within the past five years. By the end of next week, Intel expects to have issued updates for more than 90 percent of processor products intro…

Presumably the launch date, Intel lists launch dates for all of their processors on ark.intel.com [1] 5 years is Haswell, so Haswell and newer should get the microcode fix. A 3 year old computer is likely to have a CPU no more than 4 years old, but you can look it up to verify. (Both Windows and Linux can show the CPU model, and Intel's Ark site gives the release date.) [1] https://ark.intel.com/products/series/75023…

Thanks. My work computer is in the legacy section (release Q4'11), so it looks like I might be getting a new one. A shame really, it's still mostly adequate, the only thing making it age are the continued bloat from Visual Studio.

Re: Intel CEO: Patches will come to 90% of chips in the next week

#98

Earlier quoted context omitted.

Meltdown, however, is the one that results in nontrivial performance degradation in order to patch. And Meltdown specifically only affects Intel processors.

Meltdown affects some ARM prcessors as well. https://developer.arm.com/support/security-update

My response was in the context of privong stating that Intel and AMD were equivalently affected, when it is specifically Intel that suffers performance degradation in patching due to their unsafe usage of speculative execution.

Re: Intel CEO: Patches will come to 90% of chips in the next week

#99

Earlier quoted context omitted.

The more people you tell, the higher the chance of a leak. "Loose lips sink ships."

FreeBSD isn't two people in a garage, it's a foundational part of the internet developed by professionals who can be counted upon to do the right thing. Intel and Google don't have a leg to stand on: I'm super glad they found the bugs, but their disclosure has been nothing but a shitshow.

They can't be counted on anything after they publicly announced they would ignore embargo.

Re: Intel CEO: Patches will come to 90% of chips in the next week

#100
post #51

So when will chips with a fix in silicon be available?

Just buy an AMD chip right now.

How common is the Linux segfault problem on Ryzen now? Just reading about the horrible RMA procedure puts me off AMD unless they can clean up their terrible handling of this problem.
Post reply on HN