4.16.9

l4linux

i don't have much time but it is definitely on my to-do list along with a bootloader
How would that help with the line count
Oh wow . I should try plan 9 some time
No fucking shit
I never said it all goes to the bin . It would be huge
Why?

Go to bed, Tanenbaum.

stop being such an mit nigger and write your own kernel and compiler RIGHT NOW

This. even with excluding LKMs, most of the compiled code isn't in your critical path once the config is known in startup

oh look! it's this retarded thread again
OP you faggot, at least 80% of those lines are drivers and those are usually compiled as modules and only loaded as needed
and if you weren't the terminal cancer that's killing Jow Forums, you could compile your own kernel and strip everything that's not needed on your machine

Attached: 1490542536314.jpg (470x470, 70K)

I wish there was something new (development started in last 10 years), made with no mistakes of past and legacy code and with actual hardware support.

>How would that help with the line count.
depending on the configuration, it might be that only 10 out of the 20M lines of source code are actually compiled, because you only included the subset of modules that you need. It's as if you removed part of the source code.

its not that much if you disable things that you dont need

You know i sometimes wish there was a fork of linux that threw out all the crap but kept compadibility.

Have you tried 9front?
It's quite the experience. Comes with emulators, games and browsers too.

qubes

you almost described plan9, and it would end up just as bad
maybe one day you'll realize that nothing is free and Linux development is paid for by corporations who wouldn't use it if it broke compatibility every few years
slow and steady evolution tends to bring better results that "move fast and break things"

>nothing is free
As if no corporation wants innovation.

I'm not sure how that relates to anything?

>tfw knoppix (klaus knopper) is my prof in uni for software engineering

>using garbage wireless cards

I didn't make this thread because I though all the code goes in the bin, it would be huge if that happened
I was just wondering if there was something smaller .
And by the way I am using Gentoo and I configured the kernel myself.

If they want progress, there is no reason to put millions in legacy systems.
At the very least they should make new ones in parallel.

well there is smaller, but since most of Linux is drivers then you know the primary reason it's smaller
BSDs should be pretty usable on the desktop but you'll have to be careful about buying supported hardware

actually, they'd be putting way more money into reinventing the wheel, and another huge amount into porting existing applications and writing new ones for the new system
and there's practically zero guarantee the new system will be notably better in performance than the existing one, making it hard to justify the expenses and risk

it will become just as big as linux if support for things is actually added.

just compile your own linux kernel with only the features you need

Attached: Quasimoto1.jpg (770x793, 121K)

I can't. I tried to clean up the most stuff I don't need, but if I start clean, my laptop LEDs don't work.

There's only about 500k LOC in the actual kernel code, the rest is all drivers and stuff. Half a million for something as complicated as a kernel is justified.

>remove drivers you don't need
>suddenly pretty small kernel
wow

are broadcom wireless cards open-specs? didn't think so

exokernels are the future

user, you are the smartest user on Jow Forums right now.

Do you strip everything that's not needed on your machine and compile your custom kernel

This guy is not OP btw
I am