RadeonSI Now Supports Tessellation In Mesa Git

[xeekei]

It'd be interesting to see what happens when Vulkan hits. When everyone starts from scratch (theoretically). I believe Mesa have a fair chance of leading this time.

Also, since Vulkan will essentially nullify game-specific optimisations, Mesa won't need to be behind in performance either.

[adila01]

Thanks to all those involved with the implementation! You all deserve a pat on the back.

[boffo]

It'd be interesting to see what happens when Vulkan hits. When everyone starts from scratch (theoretically). I believe Mesa have a fair chance of leading this time.

Also, since Vulkan will essentially nullify game-specific optimisations, Mesa won't need to be behind in performance either.

I wonder if the LunarG work will land in mesa.

[andrei_me]

I propose a party today and when Mesa 11 release

[pal666]

2y I would guess.

OpenGL ES 3.1 will be targeted first and also Vulkan will consume significant part of devs time (maybe it already do, at least someone from Intel floss driver crew is on board of Khronos group that work on it).

your guess is obviously wrong. less than 2y have passed since release with gl 3.3 and already every of 4.x is done for 70-80-90%. i guess from now on we will see at least one gl version bump with each (3-month) mesa release

[M@yeulC]

Also, since Vulkan will essentially nullify game-specific optimisations, Mesa won't need to be behind in performance either.

Not sure about this. Threading optimisation will be useless, yes. But shader optimisations will probably be still used. However, the shaders might be more difficult to optimize, since they will be shared in binary form, this time.
I see room here for better optimized games if they share their shaders' sources with graphics vendors.

[SXX⁣]

I wonder if the LunarG work will land in mesa.

If it's don't use any part of Mesa, why would it?

[boffo]

If it's don't use any part of Mesa, why would it?

I don't know why.
LunarG is developing for the Intel gpus, if it would land in mesa maybe it could became the "new gallium". Shared code for different hardware.
With mesa I have a "out of the box experiece", something I would like to have with OpenCL, because the adoption would be much higher in the open source world. Just think about the usage of openGL 3 in open source software before mesa supported GL3.
If every company would make their proprietary versions of Vulcan, the adoption rate would not be fast.

[imirkin]

Does nvc0 stuff extrapolate to nv110 or... ?

Ssssort of. In the sense that the functionality is exposed, yes. In the sense that it works... maybe. I don't even have a maxwell, I did do basic testing on one remotely at some point for that patchset, maybe 6-8 months ago and it didn't fail any more than fermi. But there are probably subtle differences that will have to be analyzed and fixed as they are reported.

Same goes for Kepler (GK10x) and Kepler2 (GK110/GK20x). I've already discovered one weird thing about Kepler, I'm sure there are a few more lurking. On the bright side, I do have access to a GK208 now so hopefully I'll notice issues sooner.

Unlike the professional dev teams, I don't have arbitrary access to hardware, dev boxes, not to mention documentation or time to look into things. I don't push things that I know to be totally broken, but it does often happen that one or another thing doesn't work on some revision of the hardware.

[TheRealBecks]

Well... OpenGL 4.1 is ready... http://mesamatrix.net/#Version_OpenGL4.1-GLSL4.10
Waiting years for seeing OGL 4.0 getting finished and within 24 h 4.0 and 4.1 got finished? What a crazy day...