XWayland Nukes The NVIDIA EGLStream Backend

[MorrisS.]

1. Canonical didn't jeopardize anything. Nobody seriously considered their Mir option from the start, and Mir is actually a pretty cool project these days as a Wayland server.
2. NVidia also technically didn't jeopardize anything. They had a serious technical look and considered EGLStreams to be a better option.

I think many people don't realize that to this day, EGLStreams ARE the better option. The only issue is that everybody in the FOSS community had already settled on GBM since it's a part of MESA and had written a bunch of code before NVidia could argue their case. Nobody wanted to rewrite all of that code and slow down adoption of Wayland (which is hilarious looking back on it now) so they argued against NVidia. I don't like NVidia as a company, but in this particular instance I still say they were right and we spent more manpower arguing against them than it would have taken to adopt their idea.

As for "how advanced wayland would be", the VAST amount of issues with wayland are with missing protocols to do things. Nothing about GBM vs EGLStreams would have changed that. The problem is the crazy amount of bureaucracy that goes into each and every protocol added to the Wayland spec, even just "extensions" and no additions to the core spec.

I agree completely. The usual excuse is to accuse Nvidia to hide own incompetence.

[You-]

Really, this is what all the drama was about? 1.5k lines of code? The OSS community is in a bad, bad place...

Nope, it wasnt about this code. It was about it never working properly on the nvidia side, the only side that actually needed it.

[mppix]

I think many people don't realize that to this day, EGLStreams ARE the better option.
...

We could have debated these statements before a bunch of people made a very serious effort to implement EGLstreams but everybody ran into severe limitations that could not be overcome. Gnome has probably the best documented case and even Nvidia tried but nothing materialized.

I tracked this closely because I am a heavy CUDA user. Yet, I have not run a DE off a Nvidia GPU for a long time because you still need workarounds.

As for "how advanced wayland would be", the VAST amount of issues with wayland are with missing protocols to do things. Nothing about GBM vs EGLStreams would have changed that. The problem is the crazy amount of bureaucracy that goes into each and every protocol added to the Wayland spec, even just "extensions" and no additions to the core spec.

Calling BS on pretty much every part of this:
- "VAST amount of issues" .. like?; "missing protocols" .. like?
- .. bold statement that EGLstreams is on feature-parity with GBM
- "crazy amount of bureaucracy into each and every protocol" ... like? .. compared to what other industry standard/specification?

[mSparks]

Really, this is what all the drama was about? 1.5k lines of code? The OSS community is in a bad, bad place...

closer to 1400 with the required insertions.

But dont get too excited, they will probably get added back verbatim in 6 months time as part of an effort to make xwayland on nvidia GPUs more useful.

[Daktyl198]

We could have debated these statements before a bunch of people made a very serious effort to implement EGLstreams but everybody ran into severe limitations that could not be overcome. Gnome has probably the best documented case and even Nvidia tried but nothing materialized.

I tracked this closely because I am a heavy CUDA user. Yet, I have not run a DE off a Nvidia GPU for a long time because you still need workarounds.

Nobody implementing EGLStreams ever actually tried to get it onto feature parity with GBM, or solve any of it's issues because everybody in the FOSS world was already dead-set on GBM. Every implementation was purely a workaround until NVidia got in line and implemented GBM, which they stated they would do long before any EGLStream implementations went live, and as such were half-assed. Nobody took the effort to try and fix any major issues because why would they? It was a temporary solution to a temporary problem so why would they devote developer resources?

Calling BS on pretty much every part of this:
- "VAST amount of issues" .. like?; "missing protocols" .. like?
- .. bold statement that EGLstreams is on feature-parity with GBM
- "crazy amount of bureaucracy into each and every protocol" ... like? .. compared to what other industry standard/specification?

1. I think you're not reading my comment in the context of "the entire history of wayland" and instead think I'm saying that wayland still has a vast number of issues.
2. See #1
3. I never said that. I said that the GBM vs EGLStreams debate had no real impact on the actual adoption rate of Wayland, because surface protocol was never a blocking issue. Even if it took people 2 years to rewrite the entire backend of Mutter and KWin to use EGLStreams, it STILL would not have affected the adoption rate or completion rate of Wayland protocols because:
4. yes, crazy amount of bureaucracy. Have you ever taken a look through the git issues for potential wayland protocol extensions? I'm not going to say it's any more or less than any other project, but OBJECTIVELY it is the number 1 factor in the time it's taking for Wayland to fix all of the issues required for major adoption.

[NeoMorpheus]

Wayland progress could have been accelerated had NVIDIA just adopted GBM from the start.

Or Linux grew a pair and kicked them out, like how Apple did.

tired of everyone bending over for them.

[You-]

Nobody implementing EGLStreams ever actually tried to get it onto feature parity with GBM, or solve any of it's issues because everybody in the FOSS world was already dead-set on GBM. Every implementation was purely a workaround until NVidia got in line and implemented GBM...

You missed the 5 years where nvidia put its own engineering talent into action to provide the integrations... and it still failed.

I don't mind nvidia too much, you know what you get with them and by the end of this year I suspect the opensource support will be pretty good.

But nvidia users... its like the fact that they actually had to fork out money for the product made worms eat their brains. I've only seen such lack of critical thinking elsewhere by Apple users.

"I paid for it, so it must be perfect. Now the rest of the world must bow down to support me."

As for this code, I dont think we have seen the end of it yet, there are nvidia GPUs out there that do not support GBM - nvidia started to support it in the release AFTER they cut support for older gpus. For them the options are the old slow open driver that nvidia prevents from gaining reclocking support, or the old nvidia driver that only supports EGLstreams.

[Quackdoc]

Going back a decade...

damn...

[Daktyl198]

But nvidia users... its like the fact that they actually had to fork out money for the product made worms eat their brains. I've only seen such lack of critical thinking elsewhere by Apple users.

I feel like I should point out that I'm an AMD user, and have been for the entirety of my PC existence (minus my very first laptop, which had an Intel iGPU but still FOSS driver). I dislike Nvidia as a whole, and maybe I didn't keep up on it but their initial pitch for EGLStreams had a lot of really good reasons for picking it that wasn't just "we don't want to do the work to implement GBM in our proprietary driver".

I wonder how many of the implementation issues came from the fact that most compositors were already written with GBM in mind, and the EGLStream code was essentially a shim rather than a ground-up implementation.

[oleid]

I think many people don't realize that to this day, EGLStreams ARE the better option. .

What if you intend to use Vulkan?