[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: meltdown.png (182 KB, 2048x3344)
182 KB
182 KB PNG
Meltdown and Spectre exploit critical vulnerabilities in modern processors.
>>
>>64132874
Post the link
www.spectreattack.com
>>
What's spectre?
>>
>>64132874
Will the mods pin some thread to be official?
>>
>>64132923
Reading memory by creating cache-timing side effects through speculative execution of branches and indirect memory access.
Might not break OS-level security like Meltdown but can certainly breach a JS sandbox.
>>
>>64132923
A related flaw to the intel one (meltdown) affecting all CPUs from all vendors.
It's harder to execute but essentially unfixable without a hardware redesign.
>>
>>64132874
I've been following this for the last day or two, mainly because I just recently upgraded my desktop, and I'd hate for those upgrades to go to waste.

PERFORMANCE:

A couple of things that are certain so far :
- Databases that manage large amounts of files/information will have the largest dips in performance. This includes servers/cloud services, and we could potentially be impacted heavily from a consumer POV
- Personal rigs and non-powerusers will "not" noticeably be affected by performance dips when patches go around. This is mainly a result of there not being enough interaction with the CPU. As a result, delays from post-patch computation will not be sizeable.

BENCHMARKS:

Several benchmarks have been published so far documenting performance hits.
Here is the affect on a Linux system:
https://www.phoronix.com/scan.php?page=article&item=linux-415-x86pti&num=2
Several benchmarks, in particular FS Mark, show massive hits.
This, however, stands at odds with a windows benchmark conducted by ComputerBase:
https://www.computerbase.de/2018-01/intel-cpu-pti-sicherheitsluecke/#update2
This benchmark shows little to no performance dip to a windows computer post-patch, which leads many people on /g/ to speculate that tested computer does not have the patch installed. Other users have pointed out that the version the testers have used does indeed have the patch installed (https://twitter.com/aionescu/status/948427734845239298 The patch installed is based on a build since November, which indeed has the Meltdown fix). Nevertheless, users are unconvinced by the benchmarks because 1) they are clearly different from what's reported by Phoronix, and 2) They do not use similar benchmarks as those used by Phoronix.
Lastly, one user compared his GeekBench score pre and post-patch.
>>64130643
Anon's score goes from 4500 single core to 3868 single core, which is a reduction of about 15%.

Hopefully more windows users will report their benchmarks




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.