core::arch::nvptx

Function f16x2_fma

source
pub unsafe fn f16x2_fma(a: f16x2, b: f16x2, c: f16x2) -> f16x2
🔬This is a nightly-only experimental API. (stdarch_nvptx #111199)
Available on target_arch="nvptx64" only.
Expand description

Fused multiply-add, round to nearest even

https://docs.nvidia.com/cuda/parallel-thread-execution/#half-precision-floating-point-instructions-fma

Corresponds to the CUDA C intrinsics: