[ 3 / biz / cgl / ck / diy / fa / g / ic / jp / lit / sci / tg / vr / vt ] [ index / top / reports / report a bug ] [ 4plebs / archived.moe / rbt ]

Due to resource constraints, /g/ and /tg/ will no longer be archived or available. Other archivers continue to archive these boards.Become a Patron!

/g/ - Technology

View post   

[ Toggle deleted replies ]
File: 234 KB, 800x612, intshill.png [View same] [iqdb] [saucenao] [google] [report]
64111635 No.64111635 [Reply] [Original] [archived.moe] [rbt]

>>64109308 previous thread

There is evidence of a massive Intel CPU hardware bug (currently under embargo) that directly affects big cloud providers like Amazon and Google. The fix will introduce notable performance penalties on Intel machines (30-35%).

People have noticed a recent development in the Linux kernel: a rather massive, important redesign (page table isolation) is being introduced very fast for kernel standards... and being backported! The "official" reason is to incorporate a mitigation called KASLR... which most security experts consider almost useless. There's also some unusual, suspicious stuff going on: the documentation is missing, some of the comments are redacted (https://twitter.com/grsecurity/status/947147105684123649 (embed)

) and people with Intel, Amazon and Google emails are CC'd.

According to one of the people working on it, PTI is only needed for Intel CPUs, AMD is not affected by whatever it protects against (https://lkml.org/lkml/2017/12/27/2). PTI affects a core low-level feature (virtual memory) and has severe performance penalties: 29% for an i7-6700 and 34% for an i7-3770S, according to Brad Spengler from grsecurity. PTI is simply not active for AMD CPUs. The kernel flag is named X86_BUG_CPU_INSECURE and its description is "CPU is insecure and needs kernel page table isolation".

Microsoft has been silently working on a similar feature since November: https://twitter.com/aionescu/status/930412525111296000 (embed)

People are speculating on a possible massive Intel CPU hardware bug that directly opens up serious vulnerabilities on big cloud providers which offer shared hosting (several VMs on a single host), for example by letting a VM read from or write to another one.



TL;DR syscalls are being slowed by as much as 4x after the fix.

>> No.64111661

Reminder that Brian Krzanich sold ~50% of his holdings in Intel leaving him with 250000 shares which is the contractually required minimum he can hold without losing his position as CEO.

Brian sold these shares mid December of 2017.

>> No.64111680


>buy intel
>get hacked

>> No.64111685

Linux is an AMD shill, who would of thought?!
@64111635: nice try shill!!

>> No.64111689


More info here: https://www.intel.com/content/www/us/en/support/articles/000025619/software.html

Basically only vulnerable if you can flash the firmware. You can only do this with valid Intel management credentials. So its mostly a non issue.

>> No.64111698
File: 68 KB, 700x700, 1445722183259.jpg [View same] [iqdb] [saucenao] [google] [report]

its already slow and this.. what the fuck intel.

>> No.64111714
File: 162 KB, 633x900, 1506960294542.png [View same] [iqdb] [saucenao] [google] [report]


>> No.64111719
File: 53 KB, 596x643, suck2.png [View same] [iqdb] [saucenao] [google] [report]

Are you really surprised?

>> No.64111720

This has literally nothing to do with the issue in OP

>> No.64111724

You pajeetel shill, this is a SEPARATE problem from the PTI vulnerability.

In fact, this just shows how hilariously fucked intel is.

>> No.64111730
File: 35 KB, 406x364, 1454256455739.jpg [View same] [iqdb] [saucenao] [google] [report]

mfw we will witness the fall on Intel

>> No.64111738

That's the wrong completely devastating issue, baka.

>> No.64111740
File: 326 KB, 900x900, cirno lacht komisch.jpg [View same] [iqdb] [saucenao] [google] [report]

>He doesn't actually know what intel ME is

>> No.64111746
File: 552 KB, 940x672, INTEL.jpg [View same] [iqdb] [saucenao] [google] [report]

>> No.64111748

It is still under embargo retard, this is a different bug

>> No.64111763

Intel is:
[_] Finished
[_] Bankrupt

>> No.64111770
File: 110 KB, 657x539, 1511684340309.png [View same] [iqdb] [saucenao] [google] [report]


>> No.64111791

If this is bait, its fucking excellent

>> No.64111808
File: 21 KB, 680x384, (you).jpg [View same] [iqdb] [saucenao] [google] [report]


>> No.64111811

>Intel has so many vulnerabilities that shills can't even keep track of which one they're supposed to damage control at a given moment

>> No.64111874

totally nothing suspicious goin on here..nope.

>> No.64111880
File: 31 KB, 379x565, 1483795622233.jpg [View same] [iqdb] [saucenao] [google] [report]

Windows 10 users are now finally fucked with the automatic updates

>> No.64111882

>Syscall overhead roughly doubles
>Insignificant performance loss for most desktop applications, but VERY significant loss for syscall-heavy applications
>That includes GPU drivers
>Gaymers are going to flock to AMD because AMD doesn't nerf their GPU performance
>Intel's monopoly on desktop CPUs will collapse because entire segments of the market will refuse to use their CPUs
holy shit guys
this might be it

>> No.64111899

Windows 7/8 users are now finally fucked if they don't install the patch.

>> No.64111907

Time to buy cheap intel machines and just use them offline

>> No.64111915

Reminder that ARM Aarch64 is also fucked

>The performance hit from this series isn't as bad as I feared: things like cyclictest and kernbench seem to be largely unaffected, although syscall micro-benchmarks appear to show that syscall overhead is roughly doubled, and this has an impact on things like hackbench which exhibits a ~10% hit due to its heavy context-switching.

>> No.64111941

People are better off with a slower system than a vulnerable one in this case.

Potentially quadruples, I think, judging by the dude on /r/sysadmin.

>> No.64111966

> November
Does that mean it's already known that long?

>> No.64111970

This vulnerability is very severe. It has to be patched

>> No.64111992

That's before they figured out that AMD is not affected.

>> No.64112002


>> No.64112010


>> No.64112016

That's when Intel's CEO sold his stocks, so clearly.

>> No.64112022

Let's summarize guys
>shitty yields
>constantly delayed 10 nm process
>foreskin thermal paste instead of solder in the ihs
>even in their fucking xeons
>netburst 2.0 like temps
>unscallable ringbus
>massive thermal infernos in their i9 processors
>that let the PowerPC 970 and the fucking 9590 like a winter look like a candle in the fucking Minnesota
>now this horrible bug that will hamper at least 30% of the performance
AMD did nothing and its winning this year.
Hell, even they will a FX renaissance.

>> No.64112047

>all that green

>> No.64112048


>> No.64112050

>>that let the PowerPC 970 and the 9590 look like a cande flame in the fucking Minnesota's winter

>> No.64112053

They've been working on this from at least early 2017 but it turned into a serious push near the end of the year.

I think there's been studies about exploiting speculative execution feature for even longer time.

>> No.64112056

Phoronix has done benchmarks with the fix in:

>> No.64112060
File: 81 KB, 626x657, 2f7.jpg [View same] [iqdb] [saucenao] [google] [report]

MY I7 920 FINE

>> No.64112073

You forgot how they blew lots of cash on failed Atoms in smartphone and tablets, the bricked Atoms C2000, and several remotely-exploitable Management Engine bugs.

>> No.64112082

guys guys what if this is a plot to make people buy ryzen

don't believe this news until theres hard proof aka a patch and benchmarks before&after. DONT FALL FOR THIS

>> No.64112105
File: 738 KB, 1280x720, mpv-shot_[HorribleSubs] Youjo Senki - 12v2 [720p].mkv_00:23:29.367-0001.png [View same] [iqdb] [saucenao] [google] [report]

AAAAAAAAAAAAAAAAAAAAAAHHHHHHHHHHHHHHHHHHHHHH everyone running a DB or network centric application on Intel is screwed!

Intel stock will drop 20% by the end of Q1 2018. Screenshot this.
>Intel Corporation Common Stock Real Time Stock Quotes

>> No.64112110

Time to spend all my spare 3000 usd to buy some AMD stocks.

>> No.64112128

so there is no point in these threads until the update hits and we see if there is any difference at all

>> No.64112131


>that hit to storage access on the SSD
Oh boy.

>> No.64112144

>Intel stock will drop 20% by the end of Q1 2018.
Not just their stock, but also their performance.

>> No.64112148

Sauce please.

>> No.64112150
File: 64 KB, 759x746, finished-and-bankrupt.png [View same] [iqdb] [saucenao] [google] [report]

holy fuck!

>> No.64112154
File: 312 KB, 1896x1864, 1485668203527.jpg [View same] [iqdb] [saucenao] [google] [report]


>> No.64112157

Go back to Twitter, kid

>> No.64112165

Guess we know how they achieved their "single-threaded" performance now.

>> No.64112173
File: 215 KB, 800x600, secure system.png [View same] [iqdb] [saucenao] [google] [report]

Feels good to use a secure CPU.

>> No.64112181

on a secure os!

>> No.64112189
File: 58 KB, 602x635, main-qimg-e95037a5cb8f5a5e26db60f221cd6869-c.jpg [View same] [iqdb] [saucenao] [google] [report]

what about BSD?
any relevant patches anyone?

>> No.64112195

Dude that's nothing, have you seen the I/O performance? OVER 50% SLOWER, fucking insane, this is going to be a fatal blow to Intel.

>> No.64112196

Welp, looks like it might finally be the end of my 2500K.

It was fun while it lasted.

Goodbye, old friend.

>> No.64112198
File: 15 KB, 480x360, 1464841662443.jpg [View same] [iqdb] [saucenao] [google] [report]


>> No.64112202

Optane is fucking dead now.

>> No.64112204

you are forgetting about most motherboard vendors whom haven't even patched all their system affected by the intel management engine bug.
"intel management engine bug" has a good sound to it, given current events, brian is probably the exploit that management engine

>> No.64112205

Include me in the meme screenshot

>> No.64112206

I wonder how Intel will advertise their next couple generation when they fixed the bug and can't use this speculative execution anymore to improve IPC.

New Intel core i9-9870k with -10% IPC

>> No.64112207
File: 33 KB, 347x393, zpD6qre.jpg [View same] [iqdb] [saucenao] [google] [report]


>> No.64112224

I'll upgrade to some minimal distro and make an imageboard, like that dude some days ago

>> No.64112226

I summon Charlie.

>> No.64112232

I need to fucking short intel

>> No.64112233
File: 709 KB, 960x720, mpv-shot_[joseole99-TCL] BECK - 07 (960x720 h264 AC3 AC3) [89651CC3].mkv_00:17:16.951-0001.png [View same] [iqdb] [saucenao] [google] [report]


>> No.64112243
File: 67 KB, 510x332, 1500261108728.png [View same] [iqdb] [saucenao] [google] [report]

However, it may be that the vulnerability in Intel's chips is worse than the above
mitigation bypass. In an email to the Linux kernel mailing list over Christmas, AMD
said it is not affected. The wording of that message, though, rather gives the game
away as to what the underlying cockup is:

>AMD processors are not subject to the types of attacks that the kernel page
>table isolation feature protects against. The AMD microarchitecture does not
>allow memory references, including speculative references, that access higher
>privileged data when running in a lesser privileged mode when that access
>would result in a page fault.

A key word here is "speculative." Modern processors, like Intel's, perform speculative
execution. In order to keep their internal pipelines primed with instructions to perform,
the CPU cores try their best to guess what code is going to be run next, fetch it, and
execute it.
It appears, from what AMD software engineer Tom Lendacky was suggesting above,
that Intel's CPUs speculatively execute code potentially without performing security
checks. It seems it may be possible to craft software in such a way that the processor
starts executing an instruction that would normally be blocked – such as reading
kernel memory from user mode – and completes that instruction before the privilege
level check occurs.
That would allow ring-3-level user code to read ring-0-level kernel data.

>> No.64112254

If I'm not running VMs is there any reason why I would be affected by this update?

>> No.64112263

>Intel's low-powered Bonnell microarchitecture employed in early Atom processor cores also uses an in-order dual pipeline similar to P5

>> No.64112270

Why would you buy anything now wait for black friday

>> No.64112272

Yes. This flaw has nothing to do with virtualization.

>> No.64112276

Depends, is your CPU Intel and running OS?

>> No.64112281

Depends on what you do and how utilized your processor currently is.

>> No.64112283

>If I'm not running VMs is there any reason why I would be affected by this update?

The defacto loss (I can read your keys yo) of all privilege separation on all OSs.

>> No.64112285

FreeBSD is being patched as well.

>> No.64112288

There's already a patch and benchmarks. Just not from news sites.

>> No.64112290

Everything is fine if it can't go below ring 0

>> No.64112291

VMs ≠ Virtual Memory. Affects all CPUS.

>> No.64112295
File: 52 KB, 719x583, Fuck_C.jpg [View same] [iqdb] [saucenao] [google] [report]


>> No.64112308
File: 1.25 MB, 1845x1923, 1514937486656.jpg [View same] [iqdb] [saucenao] [google] [report]

>> No.64112330
File: 123 KB, 638x599, 1508959299069.jpg [View same] [iqdb] [saucenao] [google] [report]


>> No.64112331


Intel is literally going to collapse.

>> No.64112333

If I run a host with VMs, do I only have to patch the host OS or the VMs, too?

>> No.64112335
File: 27 KB, 403x401, cpu.png [View same] [iqdb] [saucenao] [google] [report]

As a more or less computer illiterate retard, what does this mean for my computer?

>> No.64112340

Fuck you, at least he was competent enough to not let shit like this happen.
I miss him. ;_;

>> No.64112341

5-30% performance hit in a few cases.

>> No.64112351

It will get slower if you install windows updates
If you don't then your system has virtually no security any more

>> No.64112361

If you like your computer, then you can keep your computer.

>> No.64112365

If I apply the patch? I haven't used windows update since mid '16.

>> No.64112371


>> No.64112372

Time to turn off windows updates for good

>> No.64112378

54% slower IO speed on NVME samsung 950 with a i7 8700k First graph


>> No.64112381
File: 100 KB, 779x1150, ssd-perf-on-insecure-x86-platform.png [View same] [iqdb] [saucenao] [google] [report]

Signed-off-by: Tom Lendacky <[email protected]>
arch/x86/kernel/cpu/common.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index c47de4e..7d9e3b0 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -923,8 +923,8 @@ static void __init early_identify_cpu(struct cpuinfo_x86 *c)


- /* Assume for now that ALL x86 CPUs are insecure */
- setup_force_cpu_bug(X86_BUG_CPU_INSECURE);
+ if (c->x86_vendor != X86_VENDOR_AMD)
+ setup_force_cpu_bug(X86_BUG_CPU_INSECURE);


>> No.64112388


If you don't apply the patch your computer will get compromized and taken over by a hacker.

>> No.64112395
File: 64 KB, 613x1247, itsafeature.png [View same] [iqdb] [saucenao] [google] [report]

It's a feature you filthy goys, you should be happy.

>> No.64112401

When can I sue the Intel/NSA?

>> No.64112402

ayymdee bros [email protected]

>> No.64112424

I already have no AV anyway

>> No.64112433

In reality this is a plot to force people to hastily adopt a rigged security update. Rigged by the merchants.

>> No.64112494

This is actually good. If you're a gaymer you get fucked, but if you waste your time in idle lurking you won't notice anything

>> No.64112502

This is only a rumour, fucking faggots

>> No.64112513

Don't you have more stock to sell, Brian?

>> No.64112517

>visit popular online hardware store
>see drop in price of Intel
>see this thread after
It all makes sense now.

>> No.64112519
File: 86 KB, 1127x685, 1488817512242.jpg [View same] [iqdb] [saucenao] [google] [report]

sounds like an obama promise

>> No.64112526
File: 944 KB, 1228x1502, 1500010756103.png [View same] [iqdb] [saucenao] [google] [report]


>> No.64112531

the security issue is a rumor, the patch is real

>> No.64112533


>> No.64112541

Guys, this is all overblown!

We can just hack the Intel ME with the vulnerabilities discovered last month and be safe in ring -3 from ring -1 vulnerabilities!

>> No.64112542
File: 124 KB, 882x731, 1481325274782.jpg [View same] [iqdb] [saucenao] [google] [report]

Intel CEO dumped as many shares as he could 2 weeks ago. Only holding onto the 250,000 he's required to keep as CEO.

>> No.64112555
File: 379 KB, 1111x597, 1451808595511.jpg [View same] [iqdb] [saucenao] [google] [report]

I work in hosted platforms and I'm dying. The funniest thing about this all, is that thousands of companies have or are switching to cloud based solutions with so like thin client + cltrix/vmware so that they could/can ditch their XP and 7 machines after shitting themselves over WannaCry.

This is gold. Pure gold.

>> No.64112558

My raspberry pi is secure
Since it can now run OpenGL, there's literally no reason to use anything but that

>> No.64112568

The fix is potentially that big of a slowdown?
Who the fuck thought that was acceptable?

>> No.64112571

>I'm dying because we're going to become bankrupt soon

>> No.64112585

sorry to break it to you, but: >>64111915

>> No.64112596

>le epic ME bug

>> No.64112599

Can you use responsive websites or are you just outta reach?

>> No.64112603
File: 17 KB, 400x400, 1488482796138.png [View same] [iqdb] [saucenao] [google] [report]

You think my laughter is joy?

>> No.64112604

There are already patches to the linux kernel, just with redacted comments.

>> No.64112605

my Raspi is 32bit

>> No.64112608

>The fix is potentially that big of a slowdown?
> Who the fuck thought that was acceptable?
kernel devs can't help it, Intel CPU's are just dog shit, don't blame them

>> No.64112614

grsecurity are a bunch of fags

The only experts grsecurity has are how to bypass the GPL and send cease and desist letters.

>> No.64112634

What is causing the huge difference between the impacts on 8700 and 6800?

>> No.64112635

When you have to beat your competitive by any means possible you tend to cut a few corners and hope they aren't noticied

>> No.64112656

The 8700 is using a SSD while the 6800 is using a HDD. The slowdowns are much noticeable with the former setup since the slow access speeds of an HDD make the processor slowdown negligible.

>> No.64112661
File: 62 KB, 630x630, sensible-chuckle.jpg [View same] [iqdb] [saucenao] [google] [report]

>2) Namespace
> Several people including Linus requested to change the KAISER name.
> We came up with a list of technically correct acronyms:
> User Address Space Separation, prefix uass_
> Forcefully Unmap Complete Kernel With Interrupt Trampolines, prefix fuckwit_
> but we are politically correct people so we settled for
> Kernel Page Table Isolation, prefix kpti_

>> No.64112663

8700k uses 950PRO, 6800k uses a SATA SSD.

>> No.64112669

CPU here is irrelevant, it's bottlenecking the SSD.

>> No.64112673

Only linux will get the slow patch. The microsoft one wont be even noticeable.
Linux wont be allowed to cheat any more.

>> No.64112683

>he actually thinks this

>> No.64112685

How on earth will datacenters, server companies deal with this? They will have instant revenue loss from the second they update and lose efficiency..

>> No.64112687

>What is causing the huge difference between the impacts on 8700 and 6800?

6800 uses a standard SATA 6GB/s SSD so only about 544MB/s not much load.

8700 uses the M.2 NVME SSD which has 3000MB/s performance.

Basically the 8700 can only handle less than half (46%) of NVME speeds now with the patch. With the regular SSD you probably won't see the difference.

>> No.64112690

Oh boy. I work in telecoms.
A lot of our customer's stable software is on intel servers, updating all of those linux kernels and back applying everything is going to be a fucking nightmare.
And then the products are handling calls, so they're performance critical.

>> No.64112692

This is what winbabbies actually believe.

>> No.64112697
File: 185 KB, 960x719, 1476201964872.jpg [View same] [iqdb] [saucenao] [google] [report]

This is it, it's the gaymer holocaust.
2018 is going to be a great year boys.
Dump all of their Intel hardware, they can't afford the huge performance hit.

>> No.64112698

You will see extremely smug Lisa Su painting in every major 7+1 CEO's office.

>> No.64112702

>>Basically the 8700 can only handle less than half (46%) of NVME speeds now with the patch
Yeah, what the fuck. Holy fucking shit, and I have 2x 980 NVMEs lol

>> No.64112705

Blame the repeal of Net Neutrality is something I can see

>> No.64112725

Sudden run to AMD hardware for servers I guess.
Maybe a lot of lawsuits against intel

>> No.64112748

I don't care about security
can I turn this off on win10?

>> No.64112752

Google has been eyeballing building their own POWER based (ppc64el) servers for years. Maybe this will get them to switch. Microsoft is intimately tied to x86 so they'll go to AMD.

>> No.64112754

God knows, my dude. Not only does the performance hit impact us, but when it is unpatched this flaw can exploit the FUCKING HYPERVISOR meaning that ALL shared and hybrid cloud services are UTTERLY FUCKED.

>> No.64112759

Probably dump performance critical system to AMD systems. Also move storage to AMD based solutions.

>> No.64112765
File: 283 KB, 900x885, 1507165656484.jpg [View same] [iqdb] [saucenao] [google] [report]

>In November Krzanich reported he had sold 245,743 Intel shares leaving him with the bare minimum of 250,000 shares required for the CEO to hold under Intel’s corporate regulations.
Really makes me think...

>> No.64112766

wasnt that a large part of a presentation on 34c3?

>> No.64112767


>> No.64112770

Enjoy the CP botnets and bitcoin miners

>> No.64112771

there is barely one rack mountable amd server to the market, no one is going to replace their fucking hardware just for this

>> No.64112782

>Meanwhile at AMD


>> No.64112786

Yes, if you upgrade to Windows 7 or 8.1

>> No.64112791


Well when your performance critical hardware suddenly drops 5-65% of perf you will.

>> No.64112800

I should have waited the 2 more months for Ryzen when I was building my PC

>> No.64112801

see, this is how sjw-free development works - fun is allowed.

>> No.64112806

>this flaw can exploit the FUCKING HYPERVISOR meaning that ALL shared and hybrid cloud services are UTTERLY FUCKED.
Sorry, hypervisor?

>> No.64112819

They are already building a POWER9 cluster (might be done now, they get prerelease hardware)

>> No.64112820
File: 2.01 MB, 2592x1936, 1345359760104.jpg [View same] [iqdb] [saucenao] [google] [report]

why has intel been in decline since sandy?

for as much as people say AMD is a "joke", and they've no doubt had mis-steps, they've also progressed a great deal in terms of actual technology and products. intel is just issue after issue while delivering weak performance "upgrades" and socket changes every year.

>> No.64112825

oh so now its 65%? no one even knows the affect it could just be 5% for 99% of uses, in which case no, you will not repurchase

>> No.64112836

The software that pretends to be hardware to VMs.
VirtualBox is one example.

Basically this bug means software in VMs might be able to access cpu memory in the host, which is a complete shitstorm for cloud services.

>> No.64112841
File: 973 KB, 801x1500, 1490523294146.png [View same] [iqdb] [saucenao] [google] [report]

Is there a list of affected cpu's? I need to make sure my bentium m shitposting machine is safe

>> No.64112843
File: 35 KB, 1024x576, 56e15f04dd0e4c75a2c3e577e942210c8c9a392d_hq.jpg [View same] [iqdb] [saucenao] [google] [report]

I have a question.

Since this vulnerability has been around since 2011, why have there been no viruses exploiting it? Why hasn't Javascript in my browser killed my computer despite me visiting dozens of shady websites?

>> No.64112853

The hypervisor is the bit of code on a VM box's host OS that manages the VMs' accesses to host system resources like CPU and memory. Amazon, Microsoft, and Google have all been involved with getting patches fixed for this ASAP.

>> No.64112857

It depends on the programme you're running.
If you don't make many syscalls it's 5%. If you're writing to disk a lot it's over 50%.

>> No.64112859

This also means the update fucks anyone trying to run services in a VM (meaning literally every major company on earth). DB servers to redis caching, all will be affected to a non zero extent.

>> No.64112860

What the fuck makes you think javascript is an attack vector in the first place?

>> No.64112864
File: 1.02 MB, 2025x9065, 1503237629957.png [View same] [iqdb] [saucenao] [google] [report]


My mates

What the fuck

>> No.64112865

Westmere and later.

>> No.64112869

Intel's core2 architecture was just recycled pentium M, which is a recycled pentium pro from 1993. All new intel shit, with the exception of atom, is still based on that.
They simply can't build anything good

>> No.64112870
File: 91 KB, 653x726, 1514919299662.png [View same] [iqdb] [saucenao] [google] [report]


>> No.64112872

>trusting kikes
That's why

>> No.64112875

You need to discover the vulnerability before using it.

>> No.64112880

My god. I use VMs but I didn't know what hv was. Not that knowledgeable. Jesus.

>> No.64112883

>citation needed

>> No.64112884

Pure I/O is 50% slower. Intel is going to be utterly banished from data centers.

>> No.64112886

because viruses not written by jews don't have a habit of notifying you that they're stealing your information. running the exploit in a way that doesnt crash your browser would be very high priority for them.

>> No.64112887

That's just what was tested. It likely affects anything since Pentium pro

>> No.64112892

A hypervisor is the virtual computer that manages the point between real metal hardware and virtual.

>> No.64112906

Is it 50% slower, or is it capped? Someone needs to break out RAID 0 and check.

>> No.64112916

You can't.

>> No.64112917

Probably the guy that submitted the KAISER patch to the Linux kernel using JavaScript as an attack vector for fiddling with KASLR timing.

>> No.64112919

Every server company is going to sue Intel dry to make up for the losses.
This is going to be glorious to watch.

>> No.64112923

This. With cloud content providers being all the rage raw I/O is of paramount importance. Intel just got fucked beyond words here.

>> No.64112927

Researchers only discovered the exploit last year apparently.
But once the embargo ends every chink and russian botnet will know about it.

>> No.64112933


>To make the transition from user mode to kernel mode and back to user mode as fast and efficient as possible, the kernel is present in all processes' virtual memory address spaces, although it is invisible to these programs.

I am dumb, are they saying the entire kernel is being copied when a new process is created?

>> No.64112934

Best part is even if I wanted to get jewed again and upgrade to Intel I have to buy a new motherboard anyway because fuck kaby lake.

>> No.64112945

see >>64112864

>> No.64112946

How many cloud operators are going to just spoof the CPU and platform strings for the guest to not lose performance?

>> No.64112950

>last year
Sounds like it.

>> No.64112951

Why is it hitting the ix-x7xx chips harder than the ix-x6xx chips?

>> No.64112955
File: 660 KB, 698x840, 1498666673445.png [View same] [iqdb] [saucenao] [google] [report]

more like 50%

>> No.64112956

Not copied around physical RAM. It's just available as part of the process's virtual memory address space.

>> No.64112965

This will fuck over all modern telecoms software bad.

>> No.64112966

One uses NVMe, the other is SATA. It's I/O testing.

>> No.64112971
File: 9 KB, 564x385, logo.png [View same] [iqdb] [saucenao] [google] [report]

I think we need a snappy logo and name for this.

This is my attempt at a logo. It's just like what all the hip marketing companies produce.

And it should be called the kneel nuclear bug because that was the first two words spat out from a random word generator.

>> No.64112974

One's using NVME SSD, the other's using SATA SSD. It's just showing how the former will be bottlenecked while the latter is less affected.

>> No.64112976

One is SATA SSD, other is NVME SSD.

>> No.64112983

it's virtual memory addresses. They all map to the same physical address where the kernel lives

>> No.64112989

That's a very real possibility, well, expect massive data leaks, 2018 keeps on giving.

>> No.64112990

The Holocaust, but real.

>> No.64112994

so everything that uses intel is fucked, it just depends how much currently

so shit like cloud service will be hit among hardest together with servers

while gaymens will only suffer the little 5%


>> No.64112996
File: 187 KB, 568x612, 1501912030011.png [View same] [iqdb] [saucenao] [google] [report]


>> No.64112997

Ahh I see it now. Thanks.

>> No.64113002

A few, until their VMs get compromised and they go bankrupt within a span of hours.

>> No.64113008

So is the Pentium with MMX affected? I'm hoping it doesn't have speculative execution when I grab one from goybay.

>> No.64113010
File: 45 KB, 704x396, linus.jpg [View same] [iqdb] [saucenao] [google] [report]


>> No.64113012

Fuuuuck Intel cucks btfo eternally, macfags btfo. I hope you like that Jew spunk on your face manbabies. Need to gloat been telling the newfags for the past 6 years.

>> No.64113013

different SSD's are used.

The 8700 uses a NVME high performance SSD 3000MB/s vs the 6800 only uses a regular SATA SSD with about 555MB/s

>> No.64113019

This will destroy almost every thin client infrastructure. This is so fucked, I am at loss for words.

I don't even want to think about DR sites and replication....

>> No.64113021

>currently under embargo
>Google shows 9001 news articles

What did he meaned by that

>> No.64113025

Gaymens may suffer a lot, since driver resides in kernel space, and your game lives in user, and they interact with fuckton of syscalls...

>> No.64113033

can i disable windows 10 updates and keep my performance? i would only use steam which in already installed

>> No.64113042

No you should upgrade to windows 7

>> No.64113050

>Still on a 3570k
>Too poor to upgrade
I'm fucked aren't I?

>> No.64113051

Yeah, but expect no security

>> No.64113058
File: 157 KB, 460x569, 12857468746845.jpg [View same] [iqdb] [saucenao] [google] [report]

Just fucking great.
My i7-3770 will be downgraded to a fucking pocket calculator and I really wanted to keep using it for another couple years.

Well done.

>> No.64113059

Just rumours that have been necromanced out of the linux code comments and patch notes, plus some security research on speculative execution from last year.
There's no official word yet or confirmation of the exact nature of the bug.

>> No.64113060

Of course, but don't mind about the bitcoin miners using about 80% of your CPU then, it's a loss/loss for Intel users.

>> No.64113061

lmao imagine how two postgres instances running in separate vms replicating to each other will behave.

there are so many unknowns with this bug its insane. intel really fucked themselves on this. the limited benchmarks being posted are showing ~5% hits in performance but they're literally all controlled tests. real world is gonna be affected by so many variables its not even funny.

..well..its funny for amd.

>> No.64113062

>macfags btfo
2018 might be the year when Apple finally moves to ARM on Macs.

>> No.64113071
File: 79 KB, 1307x333, 1497734250715.png [View same] [iqdb] [saucenao] [google] [report]

Guys there are bigger worries.

Trump is about to nuke the Gook.

>> No.64113072

would i need security if im only using steam which is already installed?

>> No.64113077

Upgrade to an FX :^)

>> No.64113081

The actual research article detailing the flaw and proof of concept exploit code has not been released. Everything is basically just speculation based off the kernel patches. The speculation is almost guaranteed correct though since making these bigass changes in an rc6 during the holidays would normally put Linus into rage mode.

>> No.64113092

Go away /pol/

>> No.64113095

>my button is bigger than yours!
jesus christ what the fuck is wrong with his brain

>> No.64113105

How serious are the lawsuits going to be?
Just the I/O performance hit is going to cause major loss to some server operators.

>> No.64113106

its running on an intel cpu

>> No.64113121

I don't want to go into work tommorow as I may go hang myself like the rest of my industry so I don't have to deal with this shitstorm.

>> No.64113123

Come on, don't tell me it's not great to have an actual madman in the office and not a pedophile ring in control.

>> No.64113130
File: 1.34 MB, 800x533, he.gif [View same] [iqdb] [saucenao] [google] [report]


>> No.64113132

How heavy were the lawsuits for the pentium bug? Or the one for the i810 chipset bug? Or the sandy bridge chipset bug?

>> No.64113134

>bug affects hypervisors and VMs.
This bad?

>> No.64113136

would i be at risk if im just using steam?

>> No.64113142

Same here man.

>> No.64113143

but does his button have an intel processor?

>> No.64113144

It's a metaphor for the US's nuclear arsenal being much more advanced and destructive than North Korea's.
Not that it's making international relations any better than they already are...

>> No.64113150

>trying to divert attention
Good goy.

>> No.64113152
File: 90 KB, 400x400, 80149182.jpg [View same] [iqdb] [saucenao] [google] [report]

KEK the FX-8370 is actually a good CPU now and will be faster than the i7-8700k is many applications after the patch

>> No.64113153

post benchmark results before you buy the rope. right now everyone is running bench tests but not actually profiling real world (since shit isn't rolled out yet). be cognizant that heavy workload users will probably start filing support tickets when their shit inevitably slows down.

>> No.64113154
File: 76 KB, 603x399, 1495997180446.jpg [View same] [iqdb] [saucenao] [google] [report]

well shit, thanks anyways

>> No.64113155

Exactly. If this wasn't critical, there would be a nuclear sized crater wherever he is from pure rage.

>> No.64113159

U.S. doesn't even need nukes to make North Korea a car park.

>> No.64113160

As soon as your computer is connected to the internet, steam or not.

>> No.64113163

Can someone write up a quick summary of this so we can use it for the next thread's OP?

>> No.64113165

They were small fry compared to this

>> No.64113167
File: 23 KB, 480x360, 1487325816952.jpg [View same] [iqdb] [saucenao] [google] [report]


>> No.64113170

Maybe not steam, but as soon as you load any site you have the potential to be infected.

>> No.64113174

Or Ryzen.

>> No.64113177

>would normally put Linus into rage mode.
I think he is in rage mode, but working hard, the rants will come later

>> No.64113178

Speculative Execution seems to start to be a thing with the Pentium Pro, which is an entirely different micro-architecture from the original Pentium and it's MMX-enhanced varriant.

>> No.64113180

>do you have Intel CPU from the last 9 years?

Short and on the point.

>> No.64113187

This article: https://www.theregister.co.uk/2018/01/02/intel_cpu_design_flaw/

>> No.64113188

Nuclear deterrence only works if the enemy thinks you are actually going to push the button. Trump is making Nuclear deterrence work better than ever before.

>> No.64113192

no hes probably in shit pants mode. the patch hes just pulled in wipes off 5% performance at a minimum from syscall heavy applications. a lot of linux niggas gonna be pissed

>> No.64113196


steam uses a browser for the shop page

>> No.64113202

I meant with a bit of detail.

>A fundamental design flaw in Intel's processor chips has forced a significant redesign of the Linux and Windows kernels to defang the chip-level security bug.

>Crucially, these updates to both Linux and Windows will incur a performance hit on Intel products. The effects are still being benchmarked, however we're looking at a ballpark figure of five to 30 per cent slow down, depending on the task and the processor model.

I'll use that, thanks.

>> No.64113203

man I hope the make the video of this guy with intel instead of nvidia

>> No.64113204

U.S. is the only country ever to use first strike deterrence. They have more legitimacy with their nuclear deterrence than any other nuclear country in the world.

>> No.64113208

anyone else notice how this """news""" fits into so many other smear campaigns? no real source, everything blown out of proportion, nothing properly explained besides "ITS OVER FOR X" and "X tards BTFO"? I've seen this pattern really come into swing since 2016

>Trump's russia thing
>Hillary's email thing
>EA's loot box thing
>some DRM thing I cant remember anymore
and now Intel. Anyone who questions it is called a shill or a Hillary voter or a numale or a cuck or a soyboy. Its getting really tiring, almost as if there is a coordinated effort to attack and anyone and everything to sow discord and unrest...

>> No.64113209


I work for a cloud hosting company. EVERYTHING is virtualized. Shared, VPS, and even dedicated servers. :( Almost all running Intel...

>> No.64113210

Sounds like every other fake news clickbait headline.

>> No.64113213

how do you know no one was ever targeted with this? do you also think no one used the nsa tools before they were leaked?

>> No.64113216

I should prob fall for the Ryzen meme but RAM prices are insane, and I'd need to buy a new heatsink as my one is pretty old and wont fit on it adding more cost. I also heard Ryzen was shit for VR but I don't get how it could be worse than my 3570k

>> No.64113218
File: 406 KB, 601x601, 1442177743317.png [View same] [iqdb] [saucenao] [google] [report]

As in infrastructure and hosting engineers, sales reps, service delivery managers, project mangers are all hanging themselves en masse.

Nah, we're finished bros. It's over. Virtualisation is dead.

>> No.64113222
File: 198 KB, 960x840, ab85cd671d05b94c4b4153735fdabf62e4c2eff321b4ea633f6c5408c8c37134.png [View same] [iqdb] [saucenao] [google] [report]

Time to sell my servers and buy just one Talos II.

>> No.64113225

>fundamental design flaw in Intel's processor chips has forced a significant redesign of the Linux and Windows kernels
>rumours that all intel CPUs from the last decade have a hardware security flaw
>linux kernel patches causing 5-65% perf hits
>worst hits to I/O
>bug allows javascript in a browser to do what it wants with your CPU, VMs can infect each other via the host
>mostly rumour so far but linus is backporting this to stable kernels over christmas without being angry so it must be big
>intel CEO sold most of his shares a month ago
>everything is fire
Use this link

>> No.64113230

nobody fucking cares about this. its nowhere near as bad as intel's fuck up.

>> No.64113231

Well then, he's fucked. kek.

>> No.64113233

Virtual machines under intel CPUs are fucked too.

Good lord what a shitshow.

>> No.64113234

I'm disconnecting my PC from the internet. I only really use it for vidya and I have my entire steam library on my HDD. I don't want that performance hit. If it turns out to be less than 10%, I will consider reconnecting it.

I hope it's not as bad as it seems.

>> No.64113237


retard why would linus be mad about a hardware problem he didnt create. Intel pays the linux foundation which pays Linus. It's literally his job to fix issues like this

>> No.64113239
File: 172 KB, 1024x768, safe.png [View same] [iqdb] [saucenao] [google] [report]


>> No.64113255

Fixing other people's figurative shit is humiliating job

>> No.64113257

It's embargoed right now, but you can work it out from the linux patches

>> No.64113261

You mean the non-existent /x/ tier bullshit? Yeah, I guess...

>> No.64113268

How its x86 emulation in POWER9?

>> No.64113270

>Trump's russia thing
>Hillary's email thing
>EA's loot box thing
>some DRM thing I cant remember anymore
Those don't affect real world business like data centers and servers, this is the end for Intel.

>> No.64113276
File: 17 KB, 342x311, cia.jpg [View same] [iqdb] [saucenao] [google] [report]


>> No.64113280


New thread, considering we'll need one in about ten minutes.


>> No.64113294

I don't know that's not me in the pic.
I will legit just use QEMU for everything if I can scrape enough crypto to switch.

Probably gonna install parabola as a server, too, over CentOS.
Freedom is non optional.

>> No.64113300

He can't rant effectively without telling everyone what the bug is, which he is not allowed to do. Expect some retroactive abortions once the embargo is lifted.

>> No.64113302


this is literally wrong, it's very rewarding to be pulled in to fix somebody else's major fuckup. It's only aggravating when you're forced to do it over and over again and they dont change their behavior and it's little things they could have avoided

>> No.64113304

There's DOSBox and QEMU.

>> No.64113305


OBSD has some mitigation from kernel relinking.

>> No.64113374

if only

>> No.64113376
File: 612 KB, 808x805, 1505371222820.png [View same] [iqdb] [saucenao] [google] [report]

how fucked am i as a random fag who just uses his computer for games and browsing the internet?

>> No.64113405

You're not at risk for security breaches but ms will slow your shit down soon

>> No.64113407

>and browsing the internet?
JavaScript can be used as an attack vector. Use NoScript with the bare minimum of whitelisting.

>> No.64113412

your games will be slower

>> No.64113413


>> No.64113415

this is just water to simmer the boil, the hardware bug cant be fixed without new silicon (and probably a good redesign of the architecture)

>> No.64113422

Feels good to be Ryzen.

>> No.64113426

With the fix;
Unknown. Gaming needs syscalls, this fucks syscalls. Bait for wenchmarks.

Without it, you're getting virus'd.

>> No.64113430

I doubt KARL can do anything against this

>> No.64113437

Nope, it beats KASLR and relinking.

>> No.64113451

Linus only gets into rage mode when one of his trusted maintainers fucks up. No-one on the team fucked up.

As usual it was the hardware company who fucked up so there is no need for direct feedback on the mailing list. Maybe at the conference when someone asks about this.

>> No.64113457

>Bait for wenchmarks.

>> No.64113461
File: 46 KB, 325x465, 1452207419414.jpg [View same] [iqdb] [saucenao] [google] [report]

>"worlds most advanced processor" i9 housefire and everything else is just a revision of a CPU from 1993

>> No.64113462

>Dump all of their Intel hardware, they can't afford the huge performance hit.

most will just move to AMD

>> No.64113471

normie from /int/ here. all memes aside, how fucked is kiketel?

>> No.64113484


>> No.64113488

Extremely fucked, based solely on the fact that a crossboarder like you is here.
Sticky when?

>> No.64113497

>>Linus only gets into rage mode when one of his trusted maintainers fucks up. No-one on the team fucked up.
>As usual it was the hardware company who fucked up

>> No.64113501

but how is the performance for the only thing x86 is worth a damn for, the bidjeogames?

>> No.64113509
File: 174 KB, 387x350, 1506485826771.png [View same] [iqdb] [saucenao] [google] [report]


>> No.64113510

nah I would prefer the pedos even if that was real

>> No.64113514

Pretty fucked.

>> No.64113520

>As usual it was the hardware company who fucked up so there is no need for direct feedback on the mailing list. Maybe at the conference when someone asks about this.

>> No.64113531

You won't be playing modern x86 shit that requires GPU access on POWER9. DOS games work very nicely in DOSBox.

>> No.64113536

on a scale of A to zyklon b they're at zyklon b.

>> No.64113540

don't you have a fucking pizza parlor with no basement to investigate?

>> No.64113541

if I/O perfromance really takes such a big hit, pretty fucked. We don't know what exactly that bug is yet, and how large the performance hit from the security patch will be. It can be everything from big 8 law departments having a field day that will make the past anti thrust shit they had to pay a few billions for look like peanuts to no legal ramifications and a few months of bad PR. At the very least they're loosing a lot of trust, which isn't a good thing for them with AMD finally pushing something competitive into the market.

>> No.64113559

>t. pedophile
You sick fucks deserve the rope.

>> No.64113566
File: 52 KB, 514x571, 1468767305181.jpg [View same] [iqdb] [saucenao] [google] [report]

>press button
>command sent to intel machine
>launch happens 50% slower
>US already destroyed at this point

>> No.64113572

>no basement

>> No.64113576

Kiketel is huge though. They can fight anything off. They can hemorrhage tens of billions and still be OK.

>> No.64113604

But can they hemorrhage tens of billions of enterprise customers?

>> No.64113615

That's not what their CEO thinks though.

>> No.64113617

sure, this won't kill them. It might cost them a lot of market share in the long run and a few billions in the forseeable future though.

>> No.64113646

>You won't be playing modern x86 shit that requires GPU access on POWER9
do they not implement PCI or PCIE?? what about the possibility of passthrough of external gpus the kind that get attached to laptops?

>> No.64113705

Passtrought can't be done because the CPU is being emulated.

>> No.64113750

Is this particular bug an isolated incident? Did they introduce it on purpose so their Chinese friends could spy on US infrastructure?

>> No.64113769
File: 123 KB, 960x960, 1512403498213.jpg [View same] [iqdb] [saucenao] [google] [report]

What if you're on Linux?

>> No.64113779

this is going to sound retarded but can't a gpu be emulated too? something like a GTX 470

>> No.64113784

Same shit.

>> No.64113795

It also will get slower.
In theory, everything can be emulated, in practice I think that it will no be that specific.

>> No.64113802

what do you mean? It affectes every single CPU they shipped out in the past decade. I doubt it was done on purpose, they surely could've found something more clever for that.

>> No.64113809


>> No.64113814

Emulation is much slower than the real thing
We can barely emulate an usable gpu in software, and then we use the whole CPU for that

>> No.64113821

of course it can. You can emulate whole CPUs/GPUs, that's how they develope new ones. Takes a few days of a huge ass server cluster to finish a few cycles though.

>> No.64113858

>in practice I think that it will no be that specific.
I thought there were already presets available to emulate model specific CPUs would assume the same would be true of GPUs
it may take till 2038 and some retarded amount of processing power but I'll finally get through my backlog of 2004-2010 games

>> No.64113885

Yes, they emulate old as fuck GPUs, and even not especific ones, but complete families.

>> No.64113897

Will Bulldozer have a sort of renaissance with this shit?

>> No.64113944

why would it? ryzen does everything better

>> No.64113974

Some people, for some reason, still are using Core 2 duos and Core 2 Quads for their shit.
I supposse that also are AMD fags that are too cheap to obtain Ryzen cpus.

>> No.64113980
File: 51 KB, 1285x420, intel.png [View same] [iqdb] [saucenao] [google] [report]

Saving this to bathe in INCT tears tomorrow morning.

>> No.64114170

The fuck is that machine. Is that a hair straightener?

>> No.64114298

i'd guess a CFL tube to replace a broken backlight. Quite impressive actually

>> No.64114743

How does it feel to be a soi boi ?

>> No.64114808

Oh god, Fallout 4 is going to be a massacre for Intel since I/O is such a massive bottleneck on that game. Running it on an SSD and using faster RAM has a noticeable impact on performance.

>> No.64115599

It does

Name (leave empty)
Comment (leave empty)
Password [?]Password used for file deletion.