The Linux Kernel Console Is Being Killed Off

Quote:

---

Originally Posted by dvdhrm

Regarding the problem with plymouth:
You cannot use plymouth together with any other application that uses graphics devices. Why do you want to use plymouth --show-splash from the console? It simply acquires all input/video devices and shows the splash screen. This doesn't work if another application (like xserver or kmscon) currently uses the devices.
I don't understand why anyone would want that? Could you elaborate a bit more?

Regards
David

---

It was for TESTING plymouth, I had just installed plymouth on my Arch box and was testing different splash screens. I just happened to run

Code:

---

plymouthd && plymouth --show-splash

---

from a KMSCON console. While I agree that you don't want to manually run plymouth and that no one would do that normally, I do feel like it is a bug that kmscon and plymouth --show-splash clash if you do want to one day have KMSCON being the full-blown replacement for the kernel VT's since you do sometimes want to test plymouth and it works just fine under the normal VT's.

Quote:

---

Originally Posted by asdx

I hope this project takes off. Please ignore all the dirty neckbeards crying over this project.

When can we start using this project? Or when it will be implemented in distros like archlinux, etc?

---

Arch has in their repositories. What I can see it work today, at lest in the limited use I tried it in.

Quote:

---

Originally Posted by Akka

Arch has in their repositories. What I can see it work today, at lest in the limited use I tried it in.

---

So how do I replace the current Linux console with kmscon?

Quote:

---

Originally Posted by asdx

So how do I replace the current Linux console with kmscon?

---

To just test it all you have to do is install it, switch a VT and run 'kmscon', to kill it switch to a seperate VT and kill kmscon. Setting it as the default VT handle is beyond my knowledge, maybe David can answer that one.

Quote:

---

Originally Posted by asdx

So how do I replace the current Linux console with kmscon?

---

Try
ln -s /usr/lib/systemd/system/kmscon@.service /etc/systemd/system/autovt@.service
systemctl enable kmscon@.service
assuming you're using systemd

OK, just over 1 MB static and stripped.
Thanks for the recommendation to try release 6, dvdrhm.
It does take some work to circumvent libtool's attempts to link everything dynamically.
And...

Code:

---

-rwxr-xr-x 1 idunham idunham 1606899 Feb 11 20:45 kmscon
-rwxr-xr-x 1 idunham idunham 1033328 Feb 11 20:46 kmscon_strip
-rwxr-xr-x 1 idunham idunham  482480 Feb 11 20:57 kmscon-test

---

That last one is what happens when I link it static with musl.
Apparently it ~works even if I stop udev.

I'm assuming "init=/bin/bash" and similar are not supported use cases? Because the only way that could work with VT emulation in userspace is if kmscon supported executing a specified command like xterm -e (for example, init=/sbin/kmscon -e /bin/bash )

Also, what does kmscon do if the xkb data is inaccessible?
OK, it fails to run.
That's ~3 MB more.

"man this code in the kernel is so hard to understand, and it's just useless bloat. let's move it to userland" where it becomes even more fragile, unmaintainable, and bloated.

sigh...

Sometimes, when you find the code you're looking at is "too hard to understand" it means that you're too stupid to muck with it, and should leave it TF alone and go do something else with your time.

Quote:

---

Originally Posted by highlandsun

Sometimes, when you find the code you're looking at is "too hard to understand" it means that you're too stupid to muck with it, and should leave it TF alone and go do something else with your time.

---

Beyond uncalled for, even the maintainer of the VT subsystem has called the code spaghetti, a mess, broken by design, and about every other phrasing you can use for shitty code. Also moving it out of the kernel is a nice idea as far as maintainability is concerned because in Userspace you can break whatever the hell you want if you need to refactor-- in the kernel once its there...its there until Linus dies.

Quote:

---

Originally Posted by highlandsun

Sometimes, when you find the code you're looking at is "too hard to understand" it means that you're too stupid to muck with it, and should leave it TF alone and go do something else with your time.

---

NO NO NO NO NO

If you have been hired by a professional corporation and you are getting paid to write the code (I presume I am talking to a basement dweller by your ignorance) then YOU ARE NOT TOO STUPID.

You have CLEARLY never actually put your fingers into someone else's complex code. Man I gotta say I would give you a good swift kick in the nads if we were speaking in person because you are just SO STUPID.

You are just UNAWARE of the software authoring process. Developers on projects come and go. Some of them leave good maintainable code behind. Some leave undocumented messes. But if you are on the team and you have resources of other developers at your disposal, you work with them to understand the odd code. And THEN after you have looked at the code and your comrades have looked at the code and you ALL throw up your hands in despair, then it is THE CODE that is the problem, not the developer.

Hey YOU can pretend that developers work in a vacuum. You can pretend that they don't work together. You can pretend that professionally hired and paid developers are "stupid". And you can GO TO HELL for your arrogant stupidity.

Quote:

---

Originally Posted by frantaylor

NO NO NO NO NO

If you have been hired by a professional corporation and you are getting paid to write the code (I presume I am talking to a basement dweller by your ignorance) then YOU ARE NOT TOO STUPID.

You have CLEARLY never actually put your fingers into someone else's complex code. Man I gotta say I would give you a good swift kick in the nads if we were speaking in person because you are just SO STUPID.

You are just UNAWARE of the software authoring process. Developers on projects come and go. Some of them leave good maintainable code behind. Some leave undocumented messes. But if you are on the team and you have resources of other developers at your disposal, you work with them to understand the odd code. And THEN after you have looked at the code and your comrades have looked at the code and you ALL throw up your hands in despair, then it is THE CODE that is the problem, not the developer.

Hey YOU can pretend that developers work in a vacuum. You can pretend that they don't work together. You can pretend that professionally hired and paid developers are "stupid". And you can GO TO HELL for your arrogant stupidity.

---

You CLEARLY don't know what you're talking about, since you have no idea how much code I have or haven't written.

You also seem to be laboring under multiple other false impressions.

1) Not everything in life is easy. Nobody said that all code will or should be easy to understand. (Just like not all code is easy to write.) The fact is, some things really are hard, period.

2) A single gifted programmer can do the work of tens of average developers. But no amount of average developers can ever do what a single gifted programmer can do.

The fact that you're professionally hired and paid says nothing about your skill. Focus on pay only says that you are a prostitute, as opposed to an actual practitioner of an art.