Experimental Linux Address Space Isolation "ASI" v2 Patches: I/O Throughput Lower By 70% (www.phoronix.com)
from petsoi@discuss.tchncs.de to linux@lemmy.ml on 11 Jan 07:44
https://discuss.tchncs.de/post/28398292

#linux

threaded - newest

M33@lemmy.sdf.org on 11 Jan 08:14 next collapse

Let’s do this Microsoft’s style: push it in production and nudge users to buy faster CPUs 🤦‍♂️

theshatterstone54@feddit.uk on 11 Jan 08:44 next collapse

The issue with that approach for the desktop is everyone will just move to other OS-es.

When Microsoft does it, you live with it cuz you have no choice.

leisesprecher@feddit.org on 11 Jan 09:15 next collapse

These patches do offer some benefits for cloud providers or in general orgs that host a bunch of different products on potentially the same machine.

I could see benefits in them, especially if the v3 or whatever addresses some of the issues.

ReversalHatchery@beehaw.org on 11 Jan 11:07 collapse

then make it default off and switchable with a kernel param, or if its a lot of code then make its compilation optional and default off

corsicanguppy@lemmy.ca on 11 Jan 09:18 collapse

Depends: did lennart poettering write it? If so, they’ll jam it down our throats.

raspberriesareyummy@lemmy.world on 11 Jan 09:52 next collapse

Thank you. Fuck that piece of shit.

M33@lemmy.sdf.org on 11 Jan 12:42 collapse

Don’t known but just to be safe let me fix it already :

system-ctl disable address-space-isolation@memory-security.service

surfrock66@lemmy.world on 11 Jan 15:08 collapse

Seems like it still in development, they have improvements in mind to reduce unnecessary system calls, and at this time you would only run these patches if memory safety was ago critical you didn’t care about IO performance, which is niche.