Re:New Spectre Vulnerability Beats Mitigations, Performance ...
by azepoi on Monday May 03, 2021 @07:01AM
(#27023571)
Is it possible to put these mitigations for Javascript only in the browser?
Re:New Spectre Vulnerability Beats Mitigations, Performance ...
by kevingadd on Monday May 03, 2021 @07:04AM
(#27023600)
Some mitigations are implemented by browsers, but a lot of it has to happen at the OS level
Re:New Spectre Vulnerability Beats Mitigations, Performance ...
by grishka on Monday May 03, 2021 @07:43AM
(#27023845)
Is it reliably exploitable from JS, or is it again a theoretical thing nobody has ever bothered to implement in practice? Is there a web page that would show me some of my kernel memory?
Re:New Spectre Vulnerability Beats Mitigations, Performance ...
by saddlerustle on Monday May 03, 2021 @08:04AM
(#27024001)
https://leaky.page/ will show you browser memory. Kernel memory is not vulnerable with browser sandboxing and current mitigations.
Re:New Spectre Vulnerability Beats Mitigations, Performance ...
by grishka on Monday May 03, 2021 @08:48AM
(#27024352)
The two tests do work for me, but the actual leaking doesn't. No matter how I tweak the parameters, it fails with "error: too many wrong false negatives in leak test (> 20%)". I guess my CPU might just be too old ¯\_(ツ)_/¯
Re:New Spectre Vulnerability Beats Mitigations, Performance ...
by 0-_-0 on Tuesday May 04, 2021 @05:00AM
(#27035423)
What browser/CPU? I only get "error: could not infer memory layout" on Firefox and Brave with a Zen 3 CPU
Re:New Spectre Vulnerability Beats Mitigations, Performance ...
by grishka on Tuesday May 04, 2021 @01:11PM
(#27039880)
Vivaldi and i7-3615QM
Re:New Spectre Vulnerability Beats Mitigations, Performance ...
by bretthoerner on Monday May 03, 2021 @08:02AM
(#27023979)
Re:New Spectre Vulnerability Beats Mitigations, Performance ...
by JeremyBanks on Monday May 03, 2021 @07:50AM
(#27023898)
It's only theoretical until we discover that some state agency got it working three years ago and has been using it to quietly pwn activists since.
Re:New Spectre Vulnerability Beats Mitigations, Performance ...
by saddlerustle on Monday May 03, 2021 @07:29AM
(#27023750)
Firefox doesn't yet, but Chrome on desktop fully mitigates spectre using site isolation - every origin (including via embedded iframes, etc) runs in its own process.
Re:New Spectre Vulnerability Beats Mitigations, Performance ...
by zinekeller on Monday May 03, 2021 @07:33AM
(#27023784)
> Chrome on desktop fully mitigates spectre using site isolation
Spectre and pals can literally jump between processes that it's no longer funny.
Re:New Spectre Vulnerability Beats Mitigations, Performance ...
by saddlerustle on Monday May 03, 2021 @08:11AM
(#27024059)
Realistically, only processes that are scheduled on sibling hyperthreads, and hyperthreading can be restricted to processes that have the same privilege level, which is done on ChromeOS. The amount of microarchetectural state that sticks around after a context switch is finite and whenever there's a new avenue found like this, usually easily fixed by ucode update and kernel fix.
Re:New Spectre Vulnerability Beats Mitigations, Performance ...
by corty on Monday May 03, 2021 @08:25AM
(#27024158)
> easily fixed by ucode update and kernel fix
Those microcode updates and kernel fixes are anything but easy. They are anything but lightweight. Context switching overhead is a huge contributor to lag and general inefficiency in interactive systems. Every kind of Spectre mitigation increases that overhead. Every further mitigation wrecks more of the performance advantages that caches have historically provided.
From a performance, power consumption and efficiency perspective, Spectre is a disaster. Mitigations just move the problem from a security and privacy one to an efficiency and performance one.
Re:New Spectre Vulnerability Beats Mitigations, Performance ...
by lights0123 on Monday May 03, 2021 @08:40AM
(#27024284)
Ironically, Firefox has done a better job at following COOP/COEP: right now, it's completely possible in Chrome (until Chrome 91, which isn't stable yet) to read other sites' JS and images without their consent with Spectre.
Re:New Spectre Vulnerability Beats Mitigations, Performance ...
by Geee on Monday May 03, 2021 @08:02AM
(#27023985)
Spectre is attack against that.
Re:New Spectre Vulnerability Beats Mitigations, Performance ...
by throwaway823882 on Monday May 03, 2021 @07:33AM
(#27023779)
Fwiw, every time Chrome claims to fix a security problem their own way, somebody finds an exploit in it later.
Spectre and the like needs to be dealt with on everything that isn't strictly an embedded fixed-code device.