[Libclc-dev] NVPTX-backend and libclc arithmetic functions

[Aaron Watry]

On Tue, Dec 16, 2014 at 5:59 AM, Jeroen Ketema <j.ketema at imperial.ac.uk> wrote:
>
> Hi Christoph,
>
>> In libclc many arithmetic functions like pow, rsqrt are lowered to
>> corresponding llvm intrinsics. These are not handled correctly by the
>> NVPTX backend.
>
> The standard llvm intrinsics do not guarantee the precision required by
> the OpenCL standard, so they should be replaced eventually. Or, only
> used for the native_ functions from the OpenCL library; in which case
> the backend should support them.
>
>> Where should this be fixed? NVPTX-Backend or libclc?
>
> libclc and there are two possible solutions:
>
> * Either completely hand-rolled versions are introduced in libclc, see
>   e.g. Aaron’s patch from yesterday.

Also, if you go the route of hand-rolling a function based on either
what AMD released recently-ish, or on the libm equivalents, make sure
that the implementation doesn't assume double support.  Many math
library implementations (e.g. libm) use doubles to calculate a more
precise float result, which means that most of the reference
implementations that are in use can't be used on things like the
Radeon 5400 I occasionally develop on.

--Aaron


> * nvptx specific intrinsics should be used that do have the correct
>   precision (if such intrinsics exist)
>
> Jeroen
>
>> Or is this expected to work and I am just not using it correctly?
>>
>> Christoph Gerum
>>
>> _______________________________________________
>> Libclc-dev mailing list
>> Libclc-dev at pcc.me.uk
>> http://www.pcc.me.uk/cgi-bin/mailman/listinfo/libclc-dev
>
>
> _______________________________________________
> Libclc-dev mailing list
> Libclc-dev at pcc.me.uk
> http://www.pcc.me.uk/cgi-bin/mailman/listinfo/libclc-dev