Looks like the missing SystemD support WILL BE a showstopper for using X.org 1.16...
 

Read and cry, my fellow die-hard Slackers!

http://lists.x.org/archives/xorg-dev...ry/040117.html

The response at question "WTF is that patch about?": X.org's DBUS is dead (i.e. kiss goodbye to plug and play mouses and keyboards), long live X.org's SystemD!*

My bet is on that SystemD support will arrive on X.org server version 1.16, which will be released on this summer...

BTW, I said that KDBUS (yep! you are right! It's about SystemD's D-BUS, at the kernel level) patches now wait on the Kernel.org RFCs?

----
* That is the sincere opinion of Hans de Goede, the patch author... :hattip:

It's a pull request. It hasn't been accepted yet.

As a rule, if the source of your news is a developer mailing list post, then it was reported prematurely. There have never been any exceptions, and there never will be. This is also one reason Phoronix is worse than useless.

Not yet, my friend, not yet...

But, I'm afraid that it hasn't been accepted yet, but will be... :hattip:

My reason is that, now, how UDEV is just a part of SystemD, direct communication with the UDEV devices via SystemD, will be an appealing solution for X.org developers, also offering support for the "modern" authentication method.

I think it's forky Mcfork fork time, I'm not letting that thing any where my X server, I don't care if I have to use free86 as a starting point and fix all those issues, atleast it won't be infected with that turd systemd.

It's a virus! And a very arrogant one...

Please...

When you want to spread FUD, at least make sure you're backed up with facts otherwise it's just... FUD.
Let's read the commit message which accompanies the "config: drop the dbus API":
What was removed? Old unused code. Good for them!

Eric

Only PV decides what's in Slackware anyway, so personally I'll just wait and see. I don't feel the need to worry about an event of which I don't know if it it will occur and when, especially if I have no way to prevent it nor to forecast its consequences ;)

After a lightning review of the patch-set, it appears an entirely optional (--enable-systemd-logind) layer. I then confirmed this by chatting with an Xorg dev.

Let's not forget Xorg still has to support non-systemd Linux, Solaris, BSDs, and others, so a hard dependency on systemd seems quite unlikely.

--mancha

Xorg can't remove dbus support or they'll destroy compatibility and piss off the entire Branded UNIX community that sponsors and financially supports most of their developments.

HP, Oracle, VMware and various other sponsors all heavily fund xorg to give proper x11 implementations. It'd be a shame to have all that funding dry up and force another fork of x11 over yet another bad decision. Plus if that funding dries up, then that would effectively kill wayland which is their biggest pending project.

If Xorg goes down, there's no tell who would pick up development, but for example if Oracle took over, chances are anything could happen, and who knows what could become of future UNIX, BSD, and Linux x11 related developments.

I'd just take the existing code, clean out all the garbage, fix security issues, and relicense it in such a way that you become BDFL of the project, if possible.

Question is for how long? I have been around for a while and even I'm starting to see things I don't like. X.org should stay away from systemd, let Wayland take the fall. Poettering has showed ignorance towards anyone who opposes his "creations", even when at a conference where he is not the guest speaker. Funny how we allow tyrants like poettering to force his creations on us, but we openly reject a murderers creation that is functional and written very well. (Hans Reiser) I don't agree for what he did, but like it or not his filesystem is hundreds of orders of magnitude better then any of Poetterings creations, Poettering makes Reiser look like a god.

Basically if your an angel in the real world you have a license to destroy Linux/UNIX's heritage, but if you did something bad in the real world your code is banished for all eternity even tho it doesn't break any fundamentals. Good call guys, if this patch makes it through I'll be reverting any systemd related code that is thrown into xorg and I will be releasing it as a fork. it's time someone says no to being force fed the "creations" (if thats what you can call them) of Poettering.

Before someone asks if I am anti-systemd YES! very much so, in fact if there were an anti-system foundation I would probly be the founder!

I agree 1000000000000000% with this article.
http://slated.org/the_poetterisation_of_gnu_linux

I'd rather be locked in a room with Hans Reiser than Lennart Poettering any day. Lennart might not make it out alive... While me and Hans have coffee and chat over file system methodology.

I honestly think more of Hans than I do Lennart, more highly and respected too. Lennart even makes dumpster scum like Bill Gates looks decent, admirable, and praiseworthy.

As pointed out by others, "the missing SystemD support WILL NOT BE a showstopper for using X.org 1.16...".

As a consequence I'd suggest that the OP mark this thread as [SOLVED] and that anyone who wants to share his or her thoughts about systemd and/or LP does so in another thread (and preferably in another forum than this one as the quota for this kind of discussion is clearly exceeded here :-)

This is not the first time that Darth Vader has posted unfounded alarmism. I have not forgotten his Future of KDE thread from several years ago.

I think too much FUD has been generated from the upstream at times that now it's starting to ooze it's way downstream and pollute things even if not implemented. Phoronix needs to full authenticate what it releases as news, or else they could be ignored for crying wolf too much.

It's too late for that. If there's a single person who isn't already ignoring Phoronix for posting outright fabrications on a regular basis, then I don't want to hear about it. Phoronix was the site that claimed that MS Office was being ported to Linux, and that Blizzard would be launching a game on Linux soon.

Blizzard has been looking into Linux, but they aren't making it top priority, same as other game publishers. There is growing interest in Linux and BSD, but nothing concrete due to the chaos going on at the moment with Linux. Right now, BSD is probably going to get more priority thanks to Sony, though I doubt PC versions of games will ever be released.

MS Office will never be ported, period. LibreOffice is already the UNIX-wide standard and the best MS Office could achieve was better compatibility through Wine and CrossOver Office.

In other Pid 1 War news, Debian has decided that...

...

...

... wait for it ...

...

...

... further discussion is needed :D

A wag on LWN has commented: "This is vastly far from my life right now (in so many ways) but for some reason I can't pull myself away, like the worst, least-gender-balanced soap opera ever. Am I alone?"

The problem is, not all distributions will even benefit from, or can even use systemd to make it usable, feasible, or worthwhile to invest the time and resources to add it in properly. Not all systems use GNOME, X, or even require dbus. Some don't even require udev.

I think Ian at Debian has a good point to ban it's usage. If it's not wanted or needed, don't force the issue.

Without reading anything, the very idea that X (X!) will twist anyone's arm is ludicrous. The sky will fall before X breaks backward compatibility, even for the sake of fixing 20 year old bugs. This is one of the major reasons a new display server had to be developed. And since it has been developed, the chances of X going anywhere are down to zero.

There is nothing at all wrong with having systemd support in Xorg, as long as it is not a hard dependency. Systemd support will help to start X as non-root user on systems that can make use of it.
Seeing that the Xorg developers are in general sane people I really doubt that they will do something like making Xorg dependent on systemd. After all, they didn't do that with Wayland either.

Doesn't X already have non-root start-up support for DRI when you invoke this in xorg.conf?

If that's the case it's nothing but a redundant feature that already is implemented.

KMS/DRI removed the need for being root regarding access to the videocard. Other parts of Xorg still run as root. Also, not all drivers support KMS.

I always run startx as regular user on Slackware. Which help do I need to do that? Do I miss something?

Yes, you miss something. Just try it, after starting X that way:

@TobiSGD: I see, thanks. But then, is that any inconvenience of having root a user of this process, or what would be the advantage of having a regular user instead?

It is a security problem. Running X as unprivileged user is far better from a security point of view.

startx runs /usr/bin/xinit which executes /usr/bin/X which is a link to /usr/bin/Xorg. And /usr/bin/Xorg is suid root.

Nonetheless, I don't see why systemd should need to get involved with how Xorg is run. Wasn't one of the main points of KMS (kernel mode setting) to obviate the need for Xorg to run as root?

It doesn't need to get involved, but it makes it easier.

Yes, it was. The only problem is: Not all drivers (just look at the proprietary Nvidia and AMD drivers) support KMS.

Xorg has a design goal of cross platform support. systemd has a design goal of being of Linux specific enhanced init system.
The premise of this thread is flawed.

systemd allows for multi-seat support under X -that's the connection. Personally, I am happy with my (one) seat...

Yeah, I can only sit on one seat at a time, only got one arse. :)

Surely there are other existing ways of accomplishing this. A trivial example is using a VM with passthrough graphics and USB hub with a keyboard and mouse, and then there's this:
https://wiki.archlinux.org/index.php/xorg_multiseat
Or do I have no idea what the heck I'm talking about, as usual?

I wish to thank all those who commented in this thread, and yes, from the different points of view, from the view of an experienced bug hunter, to opinion of one of the experienced Slackware contributors, continuing to fine analysis of the habits of the X.org developers, the response to question is definitively negative. So, I may consider this thread solved.

Thanks, Mancha, Eric, Didier and all others!