'Kernel memory leaking' Intel processor -- a serious cpu bug!?!

[bmrzycki]

Of course Intel would want to downplay the issue, and suck in all the competitors. I don't trust them one bit. Onset looks like a typical PR machination.

I don't trust Intel one bit either. Still, it's strange that the Intel patches were fast-tracked into mainline and as of today the AMD patch has not:

https://www.phoronix.com/scan.php?page= ... -4.15-Test
https://www.phoronix.com/scan.php?page= ... le-x86-PTI

[EvilDragon]

Never good being the underdog, I suppose... Intel's money talks, probably.

[jdnz]

also kontakt (and presumably omnisphere) caches sample data - which even further lowers the i/o load

Not when you're streaming hundreds of voices on the lowest DFD buffer size from your SSDs. That's a lot of random disk reads right there. Sure it might not saturate the bandwidth fully, but it IS a lot of streaming.

true - people running vienna ensemble could possibly hit high enough i/o loads for it to matter I'd expect. So potentially an issue for people working in the scoring area

[EvilDragon]

You don't even need to run VEP. Just have a bunch of Kontakt instances streaming various heavy orchestral and other libraries from multiple SSDs at a time (I have 4 1 TB SSDs just for this)...

[lnikj]

Not sure if this (detailed explanation) has been posted yet:

https://arstechnica.com/gadgets/2018/01 ... s-patches/

[bmrzycki]

Keyscape also makes _heavy_ use of streaming samples.

[cron]

Intel says it’s working with AMD and ARM in a strongly worded statement, despite AMD engineer Tom Lendacky previously saying “AMD processors are not subject to the types of attacks that the kernel page table isolation feature protects against.” Intel says it planned to disclose this issue next week along with other vendors, but that it’s issuing a statement today due to what it angrily describes as “inaccurate media reports.”

Of course Intel would want to downplay the issue, and suck in all the competitors. I don't trust them one bit. Onset looks like a typical PR machination.

Wonder if the CEO was already drafting this strongly worded statement in his head a month back as he sold every single share he possibly could. https://www.fool.com/investing/2017/12/ ... stock.aspx

Also worth noting that Intel's claim that the "average computer user" won't be affected means literally nothing. What's the average computer user? It's about as nebulous as it gets. Microsoft Office suite won't seem any slower?

[EvilDragon]

Yeah. Also, stating it "it's not a bug, but we have patches coming out soon". LOL

[teilo]

Well, I'm on macOS 10.13.2, which contains the initial patch, and I have not seen a hit at all in any DAW related work. Diva, Repro, etc. are where I would expect them to be in terms of CPU usage.

Reports from the field on the Linux patch are that I/O heavy tasks like databases (Postgres, etc.) see a hit. Games running on GL are not seeing any hit, but given how the GPU plays into things, I wouldn't expect much of a hit. More telling is that FFMPEG video encoding tasks (non-GPU boosted) are NOT seeing a hit.

So I would venture that CPU-heavy / I/O-light tasks probably are fine. I doubt we will see a hit with DAW work. Here's hoping.

[bmrzycki]

Google has confirmed this affects pretty much everyone.

https://googleprojectzero.blogspot.com/ ... -side.html

Variants of this issue are known to affect many modern processors, including certain processors by Intel, AMD and ARM. For a few Intel and AMD CPU models, we have exploits that work against real software. We reported this issue to Intel, AMD and ARM on 2017-06-01 [1].

[EvilDragon]

See CPUs tested. Zen architecture isn't mentioned, it's only older AMD CPUs, and only under a VERY specific Linux case?

[gnu23]

Intel's official statement about the issue.

[Cinebient]

Intel's official statement about the issue.

And what does that mean....is it a kind of global new world problem for nearly all cpu´s.
Even ARM mentioned...maybe i was to early to be lucky to have iOS as second music production environment.
Is this the end of all

[aciddose]

What will take "a hit" if you want to phrase it that way are cases in which process activity switches frequently between user space and kernel space.

The kernel now needs to isolate itself from user processes in order to protect itself from perverts with binoculars spying on its daughters. So it builds its mansion on top of the other side of a mountain rather than directly next to perverta village.

Now when someone from the village needs to discuss something with the kernel they need to take the long trek around and up the mountain to do so.

99% of this is total speculation (guessing, fantasy) too, so I would advise people to trust nothing until verifiable information is presented.

[camsr]

I want more details on what processors are affected. This seems to be some design flaw of out-of-order execution from what I've gathered so far, but that's not enough information to determine anything really.