Function core::arch::x86_64::_tile_dpfp16ps
source · pub unsafe fn _tile_dpfp16ps(const DST: i32, const A: i32, const B: i32)
🔬This is a nightly-only experimental API. (
x86_amx_intrinsics
#126622)Available on x86-64 and target feature
amx-fp16
only.Expand description
Compute dot-product of FP16 (16-bit) floating-point pairs in tiles a and b, accumulating the intermediate single-precision (32-bit) floating-point elements with elements in dst, and store the 32-bit result back to tile dst.