[Mesa-dev] NIR, SCons, and Gallium

[Matt Turner]

On Mon, Jan 11, 2016 at 9:53 PM, Connor Abbott <cwabbott0 at gmail.com> wrote:
> On Mon, Jan 11, 2016 at 9:21 AM, Jose Fonseca <jfonseca at vmware.com> wrote:
>> FWIW, I updated SCons to build NIR, both with GCC and MSVC:
>>
>>   http://cgit.freedesktop.org/~jrfonseca/mesa/log/?h=scons-nir
>>
>> It was actually simpler than I anticipated.
>>
>> But I hit a wall -- there's actually no way to get NIR used with
>> softpipe/llvmpipe, not even as an intermediate IR somewhere between GLSL IR
>> and TGSI, is there?
>>
>> Without this I can't actually test it.  And I'm afraid the scons integration
>> will rot again unless it is used.
>
> So, to answer the immediate question -- one way we could get NIR used
> and tested with scons/gallium relatively soon would be to pick up
> Eric's NIR -> TGSI work and use it to replace the Mesa IR -> TGSI path
> with Mesa IR -> NIR -> TGSI. It would a simpler stepping-stone to GLSL
> -> NIR -> TGSI since you don't have to deal with more modern features,
> and it would have a few tangible benefits: old assembly shaders would
> get much better optimizations before being passed to drivers, and it
> would get us closer to the goal of replacing Mesa IR entirely by
> getting rid of the second-to-last direct consumer of Mesa IR (the last
> one being r200).

Not quite true -- there's also the interpreter in swrast and i915.