[Mesa-dev] [PATCH] radeon/llvm: don't use a static array size for radeon_llvm_context::arrays

[Marek Olšák]

On Mon, Jun 8, 2015 at 5:24 PM, Jan Vesely <jan.vesely at rutgers.edu> wrote:
>
>
> On Fri, Jun 5, 2015 at 1:49 PM, Marek Olšák <maraeo at gmail.com> wrote:
>>
>> I don't think so. The LLVM IR generated by radeonsi is slightly
>> different from the one generated by r600g. The most code that can be
>> shared is shared already.
>>
>> Right now the main problem of the R600 LLVM backend is lack of
>> features. I think it's approximately at the OpenGL 3.0 or 3.1 level of
>> support. That's not even good enough for thinking about using it.
>>
>> If somebody commits to adding OpenGL 4.x support into it, maybe then
>> it will have some chance of surviving. If there are volunteers, please
>> discuss it with Dave Airlie.
>
>
> So the main work would be to add GL specific opcodes/intrinsics to the
> backend?

Yes, adding new opcodes and translating TGSI into them. The
translation part might not be straightforward, which means you'd have
to understand very well what you're doing.

Marek