[Mesa-dev] [PATCH v2 3/5] r600: implement DDIV

[Nicolai Hähnle]

On 19.01.2017 19:42, Roland Scheidegger wrote:
> Am 19.01.2017 um 18:45 schrieb Ilia Mirkin:
>> On Thu, Jan 19, 2017 at 12:40 PM, Nicolai Hähnle
>> <nicolai.haehnle at amd.com> wrote:
>>> On 19.01.2017 17:50, Ilia Mirkin wrote:
>>>>
>>>> NVIDIA can't do division either (double or otherwise - but double is
>>>> even worse since the rcp is only half there). It has some fancy
>>>> functions which get things like division, sqrt, etc.
>>>
>>>
>>> What does that mean, precisely? It might make sense to have a cap-bit after
>>> all, and then if somebody can be bothered to do it, have the front-end
>>> generate TGSI for the kind of Newton-Raphson code that LLVM implements for
>>> GCN.
>>
>> Well, there are 2 64-bit "special" (i.e. beyond
>> add/mul/fma/conversion) operations available - "rcp64h" and "rsq64h".
>> These operations consume a double, and produce the high 32 bits of the
>> result. Everything else has to be done by hand. For rcp/rsq, that
>> means newton-raphson steps, for sqrt and div, that can mean a lot
>> more, depending on how accurate you want it.
>>
>> IMHO this is something that should be left up to the drivers.
>>
>
> One problem of Newton-Raphson steps for these things is that it tends to
> screwup Inf results (which might become NaN instead), so doing it
> generically might not be worth it if hw can do better.
> (Basically the reason why llvmpipe doesn't use the sse rcp instruction
> anymore (for floats) and uses div instead: it's very fast but accuracy
> is really low - so you have to add a Newton-Raphson step. But then you
> also have to add fixups for special results, and at this point you can
> just as well use the slow division... IIRC though AVX512 actually has
> some fixup instructions which can easily replace certain values with
> Infs, Nans)

Yeah, I'm pretty much convinced that doing this generically isn't 
practically possible, and since glennk has tested the patches on cayman, 
I think we're mostly good to go. I'll wait some time to see if somebody 
wants to actually review the patches (though r600 doesn't get a lot of 
reviewing love these days...) and then go ahead with the series as-is.

Nicolai

>
> Roland
>