[Mesa-dev] Summer of Code ideas (maybe just an idea wishlist?)

[Rob Clark]

On Thu, Mar 26, 2015 at 8:59 PM, Ilia Mirkin <imirkin at alum.mit.edu> wrote:
> On Thu, Mar 26, 2015 at 8:50 PM, Rob Clark <robdclark at gmail.com> wrote:
>> On Thu, Mar 26, 2015 at 3:10 PM, Ilia Mirkin <imirkin at alum.mit.edu> wrote:
>>> On Thu, Mar 26, 2015 at 3:02 PM, Jason Ekstrand <jason at jlekstrand.net> wrote:
>>>> On Thu, Mar 26, 2015 at 12:00 PM, Aditya Avinash
>>>> <adityaavinash1 at gmail.com> wrote:
>>>>> I mean, implementing fp64 on single precision systems.
>>>>
>>>> Ok, That makes more sense!  Having lowering passes for various FP64
>>>> operations would be great.
>>>
>>> We should make sure that there are customers of such work. It only
>>> makes sense to do this when everything else but FP64 are supported for
>>> GL 4.0 (including tess). r600 (eg/ni) is an obvious customer, although
>>> that may be easier to do in sb (I believe GlennK has been looking at
>>> it, not sure how far he's gotten). Other than that... not sure. Could
>>> be that Adreno A420 can do tess && has no native fp64 support, still
>>> need to figure that out (hasn't been RE'd yet).
>>
>> fwiw, fp64 lowering is almost certainly useful for a3xx when we get to
>> the point of doing compute..
>
> Why does compute need fp64? Do the blob drivers expose the fp64 capability?
>

opencl (beyond embedded profile) requires fp64.. blob driver on a3xx
only supports embedded profile opencl (roughly everything *except*
fp64)..

>>
>> (on that topic, TGSI or NIR support for clover would be kinda awesome
>> to get compute going on some drivers other than radeon.. probably TGSI
>> would be more immediately useful for more drivers, but not sure how
>> much would need to be added to TGSI, and NIR seems to be more
>> future-proof..)
>
> Before the whole SPIR-V thing, I had been toying with the idea of
> getting clover to spit out SPIR since llvm knew how to generate it.
> Now that SPIR-V is out, I think that's a more natural target and TGSI
> replacement. But all that is somewhat in the future. Curro tried to
> get llvm to spit out TGSI, but ended up running into trouble... that
> was a few years back.

yeah, spir-v is another good option..  I would hope that by the time
freedreno's compiler backend is sophisticated enough to be useful for
compute, we have a spir-v -> nir translation of some sort.  Possibly
clover support for spir-v is the more future-proof approach to
bringing compute to more gallium drivers..

BR,
-R


>
>   -ilia