[Embench] Issues with cubic benchmark

Anton Blanchard anton at ozlabs.org
Tue Oct 22 02:19:10 CEST 2019


Hi,

> Cubic uses long double which is not standardized. On x86-64 it
> provides only 80 bits of precision, but on other architectures it
> provides either 64 bits or 128 bits of precision.
> 
> This not only means that we aren't computing the same results, but we
> are also unfairly penalising architectures that use a 128 bit long
> double. Those architectures often have a software emulation library
> for 128 bit floating point arithmetic. It was also pointed out to me
> that the benchmark isn't consistently using long double - it's using
> double versions of the transcendental functions.
> 
> I would suggest using double everywhere so that all architectures are
> doing identical computations.

A potential fix:

https://github.com/embench/embench-iot/pull/18

Thanks,
Anton



More information about the Embench mailing list