Groo (charlie@semiaccurate.com) on 4/17/11 wrote:
---------------------------
>EduardoS (no@spam.com) on 4/17/11 wrote:
>---------------------------
>>>Here is another thing we don't know about gpgpu. How *reliable* are results calculated by gpgpu?
>>
>>If your code relies on 1+1=2 then it's very reliable.
>>
>>GPUs are deterministic, all integer operations will give the same results a CPU
>>would, fp operations may differ from CPU results, but the same input will always
>>return the same result up to (I hope for a doc quoting the value) a given precision,
>>if you don't rely on the bits beyond this point it's reliable.
>>
>I know the Evergreen (5xxx and 6000-6899) and NI/SI (6900-6999) shaders are currently
>fully IEEE FP compliant with one exception, and that is a certain type of interrupt
>handling. For numerical stuff, it is complete, for the programming side, that is the one problem.

Really?

*Fully* compliant? As in <= 1/2 ULP precision guarantees for add, subtract, divide, multiply *and* sqrt?

Denormal support?

+-0 and +- infinity correctly generated and handled?

I'm very skeptical.

-- Ian.