Kernel-Based X11 Server Claims 2x Performance Over X.Org

[smitty3268]

I'm sure we'll hear about all the downsides of such a thing in this thread. But for a hard core gamer, an X server that provides this much of a performance win would be worth it.

Games already bypass the X server entirely, afaik, with the DRI2 protocol, so this is really just for desktop performance.

Anyway, the main downside for me wouldn't be possible instability, it would be the security holes.

Remember, X is network transparent, as so many people like to remind us. It would be like sticking a web browser in your kernel and expecting it to be secure.

[Milyardo]

[*]dangerously rewriting X as a kernel module

[sic] I'd like someone to explain to me how they're able to do this in-kernel without releasing the source code?

You answered you own question, it's a kernel module they can license it however they like.

[*]and to top it off, they're running this all in a VM on Windows 7.

I don't see your point, just because the guy who uploaded the youtube video is running Windows doesn't mean they always run in it VMs or Windows. Even if they did, why do you get a say in how they develop?

[intellivision]

It's honestly just insulting.

They are

• dangerously rewriting X as a kernel module
• keeping this unverified source code completely closed
• somehow not breaking the GPL by the above statement
• and to top it off, they're running this all in a VM on Windows 7.

They need to re-evaluate their priorities, and I'd like someone to explain to me how they're able to do this in-kernel without releasing the source code?

Citations needed for dangerously, relating specifically to this project. I find it interesting that you use this adjective without actually knowing the workings of the project in question.

2 & 3 are the same for any proprietary kernel module, as long as it's installed by the end user there is no distribution with GPL code, which means it doesn't come into effect.

[intellivision]

Games already bypass the X server entirely, afaik, with the DRI2 protocol, so this is really just for desktop performance.

Anyway, the main downside for me wouldn't be possible instability, it would be the security holes.

Remember, X is network transparent, as so many people like to remind us. It would be like sticking a web browser in your kernel and expecting it to be secure.

IIRC, the network transparency has been stripped out of microXwin, but that in itself might not mitigate the security risk.
I imagine this would be more suited towards offline equipment.

[ethana2]

As an Ubuntu gamer..

I'd use this if the performance is really that much better. I've got an i7-3770 and a GeForce GTX 660Ti.. I use nVidia drivers as it is and I have no objection to it. I paid good money for this rig, I want to get that value out of it. I'd prefer this be open source so I could feel, like, secure-- but hell, I could dual boot Ubuntu with a distro for this and just use the same Steam partition. Dual booting for games is fine with me, I just don't like *Windows*. Like the PS4 OS, Orbis BSD. If I could play PS4 games that way, heck yeah I'd do it. I am very interested to see benchmarks for this.

Even if no one wants to use it because it's proprietary, it's a proof of concept either way. We may as well know how much performance you can wrangle out of this. Also, when my X crashes, I restart. So does everyone else in my family who is less tech savvy, so this "Oh, but an X crash doesn't take down the whole OS...." ...entirely irrelevant from a layman's perspective, and irrelevant to me. Compiz, on the other hand.. that crashes enough that I'm glad I can just restart it.

[Vim_User]

Also, when my X crashes, I restart. So does everyone else in my family who is less tech savvy, so this "Oh, but an X crash doesn't take down the whole OS...." ...entirely irrelevant from a layman's perspective, and irrelevant to me. Compiz, on the other hand.. that crashes enough that I'm glad I can just restart it.

Have fun with that when your kernel is crashing while syncing the cache to disk or while doing system updates. Let us see if you still will just restart and everything works, when it has corrupted your filesystem or you have a system in an inconsistent state.

[halfmanhalfamazing]

I'm sure we'll hear about all the downsides of such a thing in this thread. But for a hard core gamer, an X server that provides this much of a performance win would be worth it.

Perhaps we should consider merging Xonotic into the Kernel............... I mean, for a 2x performance boost? Sure! That Jedi Knight game was recently open sourced, they should merge that into the kernel too. And how about Unigine!!!

[Delgarde]

I'd use this if the performance is really that much better. I've got an i7-3770 and a GeForce GTX 660Ti.. I use nVidia drivers as it is and I have no objection to it.

Yeah, good luck on that. A proprietary driver designed to work with the X.org driver model? And you think it'll be useful on an obscure proprietary non-X.org server?

You may also notice that there's no mention of OpenGL support in the announcement. It looks like this is just basic X11 from the late 80's - no GLX, and few if any of the other modern extensions. For example, their Compatibility page indicates it works with Gnome 2 using OpenBox WM - but not with the default Metacity, since it doesn't support the xkb extensions required. Seems to me you'd barely be able to run a desktop, never mind GPU-intensive games.

[siride]

Have any of you naysayers considered that this might be a reasonable solution in certain scenarios? It uses less memory and is faster, so it may be good on low-end kiosk-type hardware. Security could be better, but with specific use cases, may not be that big of an issue.

[Kayden]

(disclaimer: this post is my personal opinion)

I'm sure we'll hear about all the downsides of such a thing in this thread. But for a hard core gamer, an X server that provides this much of a performance win would be worth it.

This does nothing for gaming. If you look at their Performance page, you'll see that they measured using gtkperf and x11perf, which only cover 2D. Not only that, but the x11perf tests they called out are an extremely limited subset of x11perf. It basically looks like they've optimized server round trips a bit, which is nice but not terribly interesting. Common practice is to avoid costly round trips wherever possible.

I'm also skeptical that their system allows for hardware accelerated 2D drivers, such as xf86-video-intel/ati/nouveau...which would hurt performance of 2D significantly. I don't see any information to indicate that it works, at least.

So, basically, we have a closed source system...which puts a complicated system that barely even has a regression test suite directly inside the kernel...leading to a security and compatibility nightmare...all for...what exactly?