Systemd 221 Fixes Bugs, Wants Distributions To Start Shipping KDBUS

[pal666]

Here we go...Lennart's getting pretty psychotic with this now. Why should he ram kdbus down the kernel team's throats?

to make you cry, isn't it obvious?

[pal666]

http://lkml.iu.edu/hypermail/linux/kernel/1504.2/00019.html

For some reason this is the earliest post I could find in that mailing list thread, but it was already NACKed before this post was made.

not every nack is the same. for example you could nack something but who will care?

[interested]

Say in 5 years, how many end user facing applictions do you suppose won't work in that scenario? It's already difficult to not use dbus now, but with kdbus going into the kernel, how much more difficult will it become to not use it?

EDIT: My point is that you can say "remove it completely", but that's not how it's gonna be.

Yes, non-systemd distros that won't use kdbus will always have the option of removing any kdbus support from the kernel since kdbus is just a kernel module.

That some programs will start to use kdbus features is only natural. Those niche distros that won't use either D-Bus or kdbus will have to make sure that relevant userland projects supports that too, by either asking very nicely or provide the support themself.

Any choice has consequences, and if a distro chooses not to support the most widely used Linux/Unix IPC system and not even have an alternative to it, they must expect more work on their side or fewer choices in userland programs.

In practical terms, I don't think this will matter much for the non-systemd distros: A major point with kdbus is exactly that it is 100% backwards compatible with D-Bus, so few services outside those low-level ones like udev etc, have much incentive to use kdbus specific features. Daemon developers won't have to change a line of code in order to enjoy the benefits of a kernel IPC when the distro is using kdbus.

So the non-systemd distros can just continue to use the D-Bus daemon like they always had regarding most user land services.

[pal666]

Exactly this kind of behaviour is why some people dislike systemd.

some dumb people who cant understand simple things like it will work on non-kdbus kernel without issues
but nobody cares what such dumb people like or dislike

Seriously, systemd developers continue their hybris mentality. "I told you so" seems very appropriate right now.

i told you no not eat shit, why are you still eating it?

[pal666]

If they DON'T want to see kdbus ever mainlined this is the right way. Congratulations.

lol, like all mainlined features of android

[psychoticmeow]

Say in 5 years, how many end user facing applictions do you suppose won't work in that scenario? It's already difficult to not use dbus now, but with kdbus going into the kernel, how much more difficult will it become to not use it?

EDIT: My point is that you can say "remove it completely", but that's not how it's gonna be.

So you might have to do some more work to support your unusual configuration, cry me a river.

[duby229]

kdbus is totally optional in every non-contrived scenario. So distros that aren't using systemd can remove it completely from the kernel if that is what they want. It is just a kernel module after all, just like so many other kernel features.

In short, kdbus will be a kernel option forever and not a requirement for using the Linux kernel.

That systemd is going to depend on kdbus when merged is a given thing too since that has been the plan for years now.

Say in 5 years, how many end user facing applictions do you suppose won't work in that scenario? It's already difficult to not use dbus now, but with kdbus going into the kernel, how much more difficult will it become to not use it?

EDIT: My point is that you can say "remove it completely", but that's not how it's gonna be.

[interested]

As long as it remains totally optional in every possible scenario. I -don't- wan't it. If everyone is a minority then count me in too. You already know just as well as I do that it won't remain optional and that everyone will have to deal it. That's not a minority. That's everyone.

kdbus is totally optional in every non-contrived scenario. So distros that aren't using systemd can remove it completely from the kernel if that is what they want. It is just a kernel module after all, just like so many other kernel features.

In short, kdbus will be a kernel option forever and not a requirement for using the Linux kernel.

That systemd is going to depend on kdbus when merged is a given thing too since that has been the plan for years now.

[duby229]

Seemed more like a vocal minority from what I saw.. And keep in mind that it isn't uncommon for kernel developers to have a kernel-centric view of the world, without really a good picture of how userspace components interact.

Anyways, I'm not an expert on dbus or various other IPC systems out there. I do know greg-kh has spent a lot of time looking into binder/dbus/kdbus, and he is a solid/sensible kernel dev that I have a lot of respect for, so I tend to trust him more if he says kdbus is the right way forward, compared to what the rest of the peanut gallery say.

As long as it remains totally optional in every possible scenario. I -don't- wan't it. If everyone is a minority then count me in too. You already know just as well as I do that it won't remain optional and that everyone will have to deal it. That's not a minority. That's everyone.

[interested]

Here we go...Lennart's getting pretty psychotic with this now. Why should he ram kdbus down the kernel team's throats?

You are behind the news. It is Linus Torvalds who are asking the distros to turn on kdbus so it can get more exposure and real world testing before he merge it into his kernel. This is pretty common practise for stuff that goes into the kernel anyhow.

So Lennart is only doing what Linus is asking him to do. Before that, the systemd developers had gone to great length to ensure that no distro would turn on kdbus as default.

That all major distros will turn on kdbus as default when it is merged is a given thing; it is really good stuff, just the fact that you can sniff D-Bus traffic with Wireshark when using kdbus is probably reason enough for many.