no_std_compat::arch::x86_64

Function _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)
Available on x86-64 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.

Intel’s documentation