Improved fastmath code generation for trig, log, and exp/pow. #6619
RE #6619 (comment): I've taken a look through the current CUDA docs and I can't seem to find a section talking about the various options available to the
Having reviewed the Numba documentation, I've also looked at the
On the CPU target,
What do you think?
@stuartarchibald interesting suggestion. Let me digest and respond later.
As discussed out-of-band earlier, I think this is bringing Numba's behaviour into line with NVCC. An example of NVCC using a less precise cos implementation with the fast math flag can be seen in: https://github.com/gmarkall/nvcc-fastmath
I think any code that was passing
In light of the above, would you agree that the fastmath dictionary suggestion wouldn't be required now?
Also refer to documentation for fastmath in jit docstring.