[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / r / s / t / u / v / vg / vr / w / wg] [i / ic] [r9k / s4s / vip / qa] [cm / hm / lgbt / y] [3 / aco / adv / an / asp / bant / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / qst / sci / soc / sp / tg / toy / trv / tv / vp / wsg / wsr / x] [Settings] [Home]
Board
Settings Home
/g/ - Technology



Thread archived.
You cannot reply anymore.



File: benchmark-3.png (21 KB, 931x510)
21 KB
21 KB PNG
Filesystem benchmarks pre and post patch

IT'S ANUDDA SHOA
>>
Goy, remove this thread or there will be consequences!
>>
DELET THIS
>>
STOP I CANT TAKE ANYMORE
>>
Casualfag here.

I would think they would have tested this shit before releasing it into the wild.
>>
>>64126171
lmao so the 60% wasn't even a fucking meme
>>
fuck
>>
>>64126242
Nvme is getting btfo. Bottom is just an ssd
>>
File: 1502526453898.jpg (19 KB, 409x393)
19 KB
19 KB JPG
Too good.
>>
File: 1514970694618.jpg (15 KB, 456x320)
15 KB
15 KB JPG
>the absolute state of Intel
>>
>>64126242
Of course not, and it's actually 63%: https://twitter.com/grsecurity/status/948170302286172160
>>
>>64126171
Intel cucks once again BTFO
>>
What cpu lines will lose 30% of performance?
>>
File: 1501245890732.png (944 KB, 1228x1502)
944 KB
944 KB PNG
fuck xmas this late?
>>
>>64126587
All of them. First non-affected CPU is the first Pentium. Yes, the fucking first one from the nineties.
>>
>>64126587
All of them are affected. Newer processors and systems that support PCID are less affected, and older processors and no-PCID systems are more affected.

Some numbers we've seen: 50% slower I/O, 25% slower Postgres queries, 30-35% slower syscall-heavy workloads (PCID), 63% slower syscall-heavy workloads (no PCID).
>>
>>64126693
do kaby lakes have pcid? goddamit, i spent 22k on a laptop only for it to loose 30% of its power a few months later
>>
>>64126764
Anon, are you fucking retarded or were you blessed with a tiny IQ? PCID has been in practically all CPUs since Sandy Bridge. Of course a Kaby Lake CPU has it. Also, 22k for a laptop? Immediate refund, you got scammed.

Silly false flagging nigger.
>>
Imagine all those NVME servers suddenly go poof!

Poor Optane!
>>
>>64126764
>spent 22k on a laptop
Nigga fucking what. Are these Zimbabwean dollars?
>>
Is this fixable in any way with a patch or are Intel CPUs fucked forever?
>>
>>64130706
Only way is to buy the next generation Intel processors. Ones that they haven't even began to develop without this flaw yet. If you're not a retard, you'll never buy Intel again.
>>
File: 1502549012809.jpg (24 KB, 267x297)
24 KB
24 KB JPG
NOOOO! IT ISN'T TRUE! IT AFFECTS AMD AS WELL, HONEST!
>>
wait, is the patch even out yet?
>>
>>64126171

Press F to pay resbekts
>>
>>64130706
This is Intel, you already know the answer goy. Buy a new CPU and a new motherboard.

I can't wait until Intel releases their first consumer CPU without this flaw and it will require a new socket that isn't coffee lake compatible.
>>
>>64126171
what did Intel mean by this?

https://forums.aws.amazon.com/thread.jspa?threadID=269858
>>
File: 1505118149434.jpg (142 KB, 720x567)
142 KB
142 KB JPG
>tfw fell for the ryzen meme
>>
>>64126171
can i get a link?
>>
>>64126171
I'm glad I got a 10 year old CPU
>>
File: 1515007343822.jpg (166 KB, 1024x600)
166 KB
166 KB JPG
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
>>
>>64132307
That's affected too if it's Intel m8, I believe the oldest affected processors are from 1993.
>>
File: mS7OMfH.jpg (475 KB, 2592x1936)
475 KB
475 KB JPG
>>64129340
Optane btfo no doubt

Lol
>>
File: 1494281722732.jpg (27 KB, 486x486)
27 KB
27 KB JPG
>63% reduction
>>
shits fucked brehs
>>
>>64126171
JUST
>>
>>64134052
Truly the most JUST company of US
>>
File: 1470682743330.jpg (65 KB, 601x601)
65 KB
65 KB JPG
NOOOOOOOOOOO!
WE WERE PROMISED TRUE AND FAST SINGLE CORE PERFORMANCE ON ALL THREADS
>>
File: bk.jpg (354 KB, 921x865)
354 KB
354 KB JPG
>>64134525
They deserved it after all.
>>
optane now is officially a flop
>>
>>64126171
So the older it is, the better off you’ll be? Gee, I wonder Intel will “fix” this...
>>
I have a skylake i5, machine is for Netflix and gaming, am I gonna affected by this in any significant way? Is this even talking about CPUs ? I thought nvme was a hard drive thing....what's going on?
>>
>>64126171
google came out today with more information. the main exploit is called spectre which has different attack vectors for it. one is called meltdown which is the worse because it allows any program to view any process memory. the other vector attacks for it only allow you view a process memory only if you know the process you want to attack before hand. while meltdown allows for complete view of any process memory without needing to know anything about it.

google stated ALL processors on the market since the 90's are effected by spectre. all tested amd, arm, and intel processors have the exploit. but so far only intel from 2009 and above and select arm processors are affected by meltdown by the nature of intel's and arm out of order design. essentially intel and arm have processors that have a larger out of order pool that allows for a longer attack window to use meltdown type attack vector. google has stated that amd should be effected by it but their current way of exploiting meltdown isn't capable yet and with time they are sure they can get the meltdown exploit to work on bulldozer and ryzen cpu's. but at the moment its only intel processors since 2009 and select arm processors that can be effected by meltdown.

ALL amd, arm, and intel processors can be exploited by all other spectre exploits though. the only current way to fix it is to disable speculative out of order processing at the processor, hardware level.
>>
>>64137380
Is my master system ok? I can't live without Alex Kidd.
>>
File: 1507179268167.gif (997 KB, 500x270)
997 KB
997 KB GIF
>>64137380
AMD says otherwise about Meltdown ever working on their CPUs... nice troll though.
>>
Are you ready to buy the next generation Intel processors (TM) that offer a brand new, never before seen feature of not getting fucked in the ass by this performance loss?
>>
File: 1430775192503.jpg (337 KB, 867x848)
337 KB
337 KB JPG
>>64137380

>main exploit
>spectre

dohoho

sounds like youre having a MELTDOWN mr shill
>>
>>64137416
https://security.googleblog.com/2018/01/todays-cpu-vulnerability-what-you-need.html
>We have empirically verified the vulnerability of several Intel processors to Spectre attacks, including Ivy Bridge, Haswell and Skylake based processors. We have also verified the attack’s applicability to AMD Ryzen CPUs. Finally, we have also successfully mounted Spectre attacks on several Samsung and Qualcomm processors (which use an ARM architecture) found in popular mobile phones.
>>
>>64137472
>Spectre
>Meltdown
Can you read?
>>
File: JUSTEL.png (127 KB, 291x288)
127 KB
127 KB PNG
>>
>>64126631
Not the first one, Pentium through to Pentium with MMX is fine.
Pentium Pro and children are all vulnerable.

pre-2013 Atoms also aren't vulnerable since they lack OoO and branch prediction entirely.
I'd imagine Quark isn't vulnerable either - not like anything uses it anyway though.
>>
>>64137531
https://googleprojectzero.blogspot.fi/2018/01/reading-privileged-memory-with-side.html
> Before the issues described here were publicly disclosed, Daniel Gruss, Moritz Lipp, Yuval Yarom, Paul Kocher, Daniel Genkin, Michael Schwarz, Mike Hamburg, Stefan Mangard, Thomas Prescher and Werner Haas also reported them; their [writeups/blogposts/paper drafts] are at:
>Spectre (variants 1 and 2)
>Meltdown (variant 3)
there are three variants, meltdown is variant 3 of the exploit. spectre is variant 1 & 2 of the exploit.
>>
>It’s worth pointing out that not only does Windows have KPTI/KVA shadowing enabled for AMD processors as well, it even has specialized shadow system call entry stubs for AMD vs Intel. This either suggests they know how to embargo properly or that Tom’s PR is not entirely accurate
https://twitter.com/aionescu/status/948576989622882304
microsoft is enabling it on amd cpu's.
>>
>>64137588
>Cortex-A75 is just as bad as Intel CPUs
rip ARM, there goes that lucrative server market
>>
>>64137588
And AMD isn't affected by Variant 3 nor 2, learn to read.
>>
>>64126171
What is FS-Mark? What is this graph even measuring?
>>
>>64131341
Nope
>>
>>64126233
>implying they are freaking out and being yelled at by their bosses so hard they're just trying to scramble and throw something out there before their asses are fried.
>>
File: 1429317637726.jpg (401 KB, 1043x1490)
401 KB
401 KB JPG
>Just bought a fancy NVMe SSD recently
Fucking hell, I treat myself for once and then this happens. Why can't I have nice things?
>>
i5 3570K

how fucked?
>>
>>64138482
Should buy some lube
>>
>>64138591
i already have some

im guessing this is more of a "skip the water based" and go straight for the petroleum jelly?

well, guess im buying AMD soon
>>
>>64126210
>>64126218
>>64126189

What is this graph?
>>
>>64138434
Post some before and after benchmarks
>>
>Just bought intel because AMD sucks with older games
I get this is the reward I get for my computer working on my first try
>>
cant they just fix this with microcode?
why are kernels patching hardware all the sudden
>>
Why does "fix" not affect my compile times on Linux 4.14?

You guys are like plebbit when they found out he had two scoops of ice cream.
>>
File: 1512813771014.jpg (55 KB, 960x540)
55 KB
55 KB JPG
>POST YFW THE INTLEL SHILLS ARE ALREADY WORKING OVERTIME IN THE FIRST WEEK OF 2018
>PIC RELATED ARE THE SHILLS WE'RE DEALING WITH
HAAAAAAAAAAAAAAAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHAHA
>>
>>64138639
literally download the fucking patch yourself dumbass and stop falling for AMDrones bullshit
>>
>Intel

HAHAHAHAHAHAHAHAHHAHAHAHA
>>
>>64138761

it can't be fixed like that. it's a architecture problem. the only fix is a software update (which causes the performance hit) or buy a chip that doesn't have the issue - which may be a few years.
>>
>>64138628

every Intel chip for the past decade is impacted so yes, you fucked
>>
>>64138744
Is the update available for Windows 8.1? For some reason I thought it wasn't until Patch Tuesday. I'm not going to install any bullshit benchmark software, but I might run a 7-zip benchmark.
>>
>>64138770
I'm asking what the graph is indicating you arse, What is it measuring?
>>
>>64139025
Filesystem speed.
>>
File: DSplNWXV4AABlgY.jpg large.jpg (117 KB, 1852x1166)
117 KB
117 KB JPG
also have a look a this >>64138966

cpu util change after patch is applied
>>
>>64126764
Did you mean several dozen laptops?
>>
>>64137586
>pre-2013 Atoms
What about Mobile Celerons in shitty old netbooks?
>>
>>64137586
Do you know the processor model numbers or name of the atoms that lack OoO?
>>
First of all stop with the false flagging.
Nothing will happen to your video games.
Nobody cares this does not affect you or any other consumer.
>>
>Consider the figures for personal computers alone: there are 1.5 billion in use today (desktop and laptop combined) and around 90% are powered by Intel chips, IDC estimates
>>
>>64126171
is the patch allready out for linux? standard arch repositories?




Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.