RadeonSI Gallium3D In Mesa 19.1 Enables Parallel Shader Compile Support

[phoronix]

Phoronix: RadeonSI Gallium3D In Mesa 19.1 Enables Parallel Shader Compile Support

Marek Olšák of AMD kicked off the new month by finally landing support for the ARB/KHR extensions around parallel shader compile support...

http://www.phoronix.com/scan.php?page=news_item&px=RadeonSI-Lands-parallel_shader

[brent]

AFAIR shader compilation is mostly asynchronous with radeonsi already. That is, shaders are compiled in separate worker threads and the main thread only blocks if the OpenGL application wants to use the shaders before they've finished compiling.

I think the number of threads that radeonsi uses is rather conservative. This extension now allows programs to increase the number of threads, for instance for a level loading phase. And of course you can actually check progress (without blocking).

[andrei_me]

What is the expected benefit of this changes?

[smitty3268]

What is the expected benefit of this changes?

GL 4.6 conformance?

[FireBurn]

GL 4.6 conformance?

These extensions don't appear to be part of OpenGL 4.6 though

[nuetzel]

These extensions don't appear to be part of OpenGL 4.6 though

https://community.khronos.org/t/official-feedback-on-opengl-4-6-thread/76018

[-]
In addition to the above features being added to OpenGL 4.6, the following are being released as extensions:

[ul]
[li]GL_KHR_parallel_shader_compile allows applications to launch multiple shader compile threads to improve shader compile throughput[/li][li]WGL_ARB_create_context_no_error and GXL_ARB_create_context_no_error allow no error contexts to be created with WGL or GLX that support the GL_KHR_no_error extension[/li][/ul]
The OpenGL 4.6 specification can be found at https://khronos.org/registry/OpenGL/index_gl.php 27.
[-]

[StandaSK]

Well, according to mesa/docs/features ( https://cgit.freedesktop.org/mesa/me...s/features.txt ) it isn't part of OpenGL 4.6.

Either that or I'm horribly misinterpreting that page (I'm quite a noob at this).

[andrei_me]

to improve shader compile throughput

Faster load times as benefit then?