Intel Confronts Potential ‘PR Nightmare’ With Reported Chip Flaw • Hacker News Board

Re: Intel Confronts Potential ‘PR Nightmare’ With Reported Chip Flaw

#491

Post by anarazel » Wed, Jan 03, 2018, 11:24 PM UTC

eropple wrote: post #488

anarazel wrote: post #465

Earlier quoted context omitted.

> I've seen that message. It acknowledges the same problems: do-nothing problems over a local unix socket. The first set of numbers isn't actually unrealistic. Doing lots of primary key lookups over low latency links is fairly common. The "SELECT 1" benchmark obviously was just to show something close to the worst case.

> The first set of numbers isn't actually unrealistic. Doing lots of primary key lookups over low latency links is fairly common. Latency through loopback on my machine takes 0.07ms. Latency to the machine sitting next to me is 5ms. We're actually (and to think, today I trotted out that joke about what you call a group of nerds--a well, actually ) talking multiple orders of magnitude through which kernel traps are be…

> Latency through loopback on my machine takes 0.07ms. Latency to the machine sitting next to me is 5ms.

Uh, latency in local gigabit net is a LOT lower than 5ms.

> We're actually (and to think, today I trotted out that joke about what you call a group of nerds--a well, actually) talking multiple orders of magnitude through which kernel traps are being amplified.

I've measured it through network as well, and the impact is smaller, but still large if you just increase the number of connections a bit.

Re: Intel Confronts Potential ‘PR Nightmare’ With Reported Chip Flaw

#492

Post by lev99 » Wed, Jan 03, 2018, 11:39 PM UTC

maxander wrote: post #483

dijit wrote: post #454

Earlier quoted context omitted.

I have a power dependent workload that scales horizontally and is currently already dominated by the cost of system calls. This will effectively, directly cause me to buy 30% more compute on a huge infrastructure. (2,000~physical machines. Quite beefy dual socket machines with a lot of memory) I know I’m not alone. Then again. Think of microservices, Kubernetes for instance; Network requests are system calls.

Will you be buying Intel-based machines? Or will you be running a hybrid-architecture cluster now? I don’t know very much about computing on that scale, but I wonder if all the people selling off Intel stock are thinking this story through.

AMD server CPUs currently outperform Intel on some multi-threaded benchmarks. This usually isn't a problem for people buying for peak-performance because you can always buy more CPUs to increase parallel programming speeds, but it's harder to make single threads faster.

It's possible that the patches applied to fix this bug will cause some single-threaded benchmarks to change from Intel being the fastest to AMD being the fastest.

Re: Intel Confronts Potential ‘PR Nightmare’ With Reported Chip Flaw

#493

Post by laythea » Wed, Jan 03, 2018, 11:56 PM UTC

matt4077 wrote: post #443

laythea wrote: post #375

When pondering whether I should be grumpled about this situation or accept it as "just how it goes", I wonder if Intel would be OK with me giving them up to 30% less money after agreeing to 100% before walking out the Intel shop without prejudice.

I don't know if their will be a recall/class-action lawsuit/whatever. But clearly there is a difference between making a mistake in what must be one of the most complicated consumer products on the one hand, and intentionally violating the terms of an agreed-upon contract? Tl/DR: Intent matters.

Good point and agreed. This will cost them, and clearly was a mistake.

Re: Intel Confronts Potential ‘PR Nightmare’ With Reported Chip Flaw

#494

Post by fleitz » Thu, Jan 04, 2018, 12:04 AM UTC

rbanffy wrote: post #270

dsign wrote: post #19

This is a clusterf /big deal. Beyond the security implications, it means that all companies paying for computing resources will have to pay roughly 30% more overnight on cloud expenses for the same amount of CPU, assuming that they can just scale up their infrastructure. I know that bugs happen and that there was nothing intentional on this one, but at times like this is hard to held at bay the temptation of claiming…

It's a good thing CPU is fairly compressible. Unless you meter it very carefully, you'll see the performance hit and it'll not impact you that much. Very few of my physical boxes are over 70% CPU utilization on a daily average. It's, however, really bad if you sell CPU cycles for a living. You just lost between 5 and 30% of your capacity. If you have a large building, you just lost part of your parking lot to the Int…

Why is it bad? Customers are buying 30% more CPU ;)

Re: Intel Confronts Potential ‘PR Nightmare’ With Reported Chip Flaw

#495

Post by noxecanexx » Thu, Jan 04, 2018, 12:48 AM UTC

jrochkind1 wrote: post #361

Best summary I've found for the somewhat technical but not hardware-or-low-level-hacker reader is arstechnica. https://arstechnica.com/gadgets/2018/01/whats-behind-the-int...

My head is still spinning writing an OS is a BIG DEAL!!!!

Re: Intel Confronts Potential ‘PR Nightmare’ With Reported Chip Flaw

#496

Post by rogerthis » Thu, Jan 04, 2018, 1:06 AM UTC

rbanffy wrote: post #463

dijit wrote: post #454

Earlier quoted context omitted.

I have a power dependent workload that scales horizontally and is currently already dominated by the cost of system calls. This will effectively, directly cause me to buy 30% more compute on a huge infrastructure. (2,000~physical machines. Quite beefy dual socket machines with a lot of memory) I know I’m not alone. Then again. Think of microservices, Kubernetes for instance; Network requests are system calls.

If your workload has no code that's untrusted, you can safely skip this patch or disable it on boot. If not, at 2000+ physical machines, it may be worth to move some of that into kernel modules that would collapse a couple syscalls into a single higher level one.

But then you have to release the module as GPL, no?

Re: Intel Confronts Potential ‘PR Nightmare’ With Reported Chip Flaw

#497

Post by Iv » Thu, Jan 04, 2018, 1:09 AM UTC

dsign wrote: post #19

This is a clusterf /big deal. Beyond the security implications, it means that all companies paying for computing resources will have to pay roughly 30% more overnight on cloud expenses for the same amount of CPU, assuming that they can just scale up their infrastructure. I know that bugs happen and that there was nothing intentional on this one, but at times like this is hard to held at bay the temptation of claiming…

More than the lawsuit, it attacks one of the core aspect of Intel's brand: performances. Intel chips are supposed to be faster. Now they are suddenly 30% slower because they carelessly implemented performance features over security ones.

Re: Intel Confronts Potential ‘PR Nightmare’ With Reported Chip Flaw

#498

Post by 0xTJ » Thu, Jan 04, 2018, 1:39 AM UTC

rbanffy wrote: post #270

dsign wrote: post #19

This is a clusterf /big deal. Beyond the security implications, it means that all companies paying for computing resources will have to pay roughly 30% more overnight on cloud expenses for the same amount of CPU, assuming that they can just scale up their infrastructure. I know that bugs happen and that there was nothing intentional on this one, but at times like this is hard to held at bay the temptation of claiming…

It's a good thing CPU is fairly compressible. Unless you meter it very carefully, you'll see the performance hit and it'll not impact you that much. Very few of my physical boxes are over 70% CPU utilization on a daily average. It's, however, really bad if you sell CPU cycles for a living. You just lost between 5 and 30% of your capacity. If you have a large building, you just lost part of your parking lot to the Int…

Another consideration is power usage in data centers. Server power usage is annoyingly complex, and once you get above 70% utilization power usage may go up considerably.

Re: Intel Confronts Potential ‘PR Nightmare’ With Reported Chip Flaw

#499

Post by mastazi » Thu, Jan 04, 2018, 1:43 AM UTC

I am confused, as far as I know two different vulnerabilities have been discovered, Meltdown and Spectre; while the first one affects only Intel CPUs, the second one affects AMD and ARM as well. So how come I'm not seeing much talk around about the latter? Is it because it is harder to exploit?

I didn't have a chance to read the 2 papers so I would appreciate a TL;DR.

I am a SW developer working with high level languages; security and OS development are not my specific fields so while I don't need an ELI5 I would appreciate a sufficiently "layman's terms" explanation.

Re: Intel Confronts Potential ‘PR Nightmare’ With Reported Chip Flaw

#500

Post by asah » Thu, Jan 04, 2018, 2:03 AM UTC

rdl wrote: post #412

I wonder how Intel will deal with this. If I were Intel, I'd offer free replacements for at-par performance, and potentially tiny cash payment for upgraded performance. Assuming the marginal cost to produce chips, especially older/slower ones, is very low, the only real cost to them is losing out on potential upgrade sales which would have happened organically, for a while. However, doing this keeps Qualcomm/ARM and…

rofl, you don't simply upgrade chips installed in 1+B servers and laptops and network devices and embedded systems. In fact, statistically-few devices-containing-a-CPU are designed to ever have their CPUs replaced. At the last, you replace the motherboard, which requires coordination with the vendor. Consider cars - good luck coordinating that recall!