What is the preferred way of configuring floating-point precision on dpcpp?
I have previously found a macro for configuring this from SYCL, but I'm unable to find the documentation for it now.
My device does not support double fp, and I want it disabled.
Thanks for reaching out to us.
Please refer to the below link for configuring floating-point operations.
>>I have previously found a macro for configuring this from SYCL
Could you please let us know which macro you found from SYCL?
You can disable the double fp using the below pragma
#pragma OPENCL EXTENSION cl_khr_fp64 : disable
Thanks & Regards,