The MINIX pill

Have you taken the MINIX pill? Now that Linux has been pozzed by a CoC, why aren't you using Minix?
>no CoC
>microkernel
>self healing
>supported by Intel
What are you waiting for?

Attached: Minix3.jpg (1125x1505, 300K)

Other urls found in this thread:

youtu.be/vlOsy0PZZyc
youtu.be/jiGjp7JHiYs
en.wikipedia.org/wiki/Tanenbaum–Torvalds_debate
youtube.com/watch?v=_TkvthH2oc8
youtube.com/watch?v=86_BkFsb4eI
twitter.com/AnonBabble

but everyone is already using it.

don't you know what Intel ME is?

It's already running in ring -3 on my CPU.

I meant in userspace

Brainlet here, what is self healing?

Attached: brainlet_feels.png (645x773, 11K)

I want to give it a try but I dunno where to start

If part of the kernel crashes, the entire thing doesn't go down, and can safely be isolated and restarted
Start with a VM and install some software

>install some software
which will fail to compile almost always.

Don't forget, utterly useless.

youtu.be/vlOsy0PZZyc
youtu.be/jiGjp7JHiYs

Its just as useful as NetBSD is

You forgot, useless.

try compiling firefox from ports on MINIX3.

we are all using minix, cianigger

You mean Intel ME runs an operating system?

>supported by Intel
ugh no, fuck you

ast is so cute, his books are fun to read

this, I bought older copies of Computer Networks and the Minix book, and both are fantastic. So well written.

Attached: 1537721501416.png (600x355, 173K)

>Firefox
Found your problem

Yeah x86 is awful

Ok, what browser does work? The most sophisticated one I could get running was w3m, and you can't exactly render much of the web with that.

the official git repo says that the latest commit that wasn't bringing the repo up to speed with the latest NetBSD release was years ago. Did they just stop adding shit to the kernel or what? Even if it's only shit like Meltdown and Spectre mitigations, I'd like to see something done to the kernel more recently than 2016.

Drivers run in user space and the kernel has a loop that regularly checks if the drivers are doing ok and if not it restarts them. The name sounds more interesting than what really is going on.

There's nothing to add

Why would I do that? Why does it matter if the drivers are not working? Wouldn't it also make the OS slower because you are running drivers in user space constantly checking if it is okay?

Imagine making a banter so epic there's a dedicated normies wiki page for it

link?

en.wikipedia.org/wiki/Tanenbaum–Torvalds_debate

Microkernel architecture does not have performance as an aim.

Minix is not copyleft and it's likely intel maintains an internal fork they fill with their own improvements and nasties that they then peddle onto the consumer without their knowledge.

>Wouldn't it also make the OS slower because you are running drivers in user space constantly checking if it is okay
Actually no, syscalls are fucking expensive in monolithic kernels due to context switching. Microkernels still have the IPC overhead but on average it doesn't take a significant additional time.

The myval pill

thread theme
youtube.com/watch?v=_TkvthH2oc8

Attached: Screen Shot 2018-09-24 at 21.33.48.jpg (2070x1128, 226K)

Virgin Linus still receives humiliation from Chad Tanenbaum even after these year for being able to self-heal kernel

>loonix runs on a bunch of servers and largely irrelevant gentoo rice machines
>MINIX runs on every single recent Intel x86 chip there is

so Tanenbaum won in the end?
en.wikipedia.org/wiki/Tanenbaum–Torvalds_debate

>jew
>chad
ok

gb2pol

Chrome

No, you.

This guy is cool as fuck
youtube.com/watch?v=86_BkFsb4eI

What about hybrid kernels like Dragonfly BSD or Plan 9?

>supported by Intel
enough of a reason to not use it

Hybrid kernel is just a marketing term

>doesn't know about L4 microkernel
On L4 with Wombat (a Linux interface running in L4 userland) context-switching costs are 50 times lower than in native Linux.

basically when your driver crash, it just restart that part, like in windows

tranny's CoC obviously better.

implying Tanenbaum wasn't at fault.

sneaky jew.