Function core::arch::x86_64::_mm256_maskz_cvtneps_pbh[][src]

pub unsafe fn _mm256_maskz_cvtneps_pbh(k: __mmask8, a: __m256) -> __m128bh
🔬 This is a nightly-only experimental API. (stdsimd #48556)
This is supported on x86-64 and target feature avx512bf16,avx512vl only.
Expand description

将包装的单精度 (32-bit) 浮点元素在 a 中转换为包装的 BF16 (16-bit) 浮点元素,并使用 zeromask k 将结果存储在 dst 中 (当相应的掩码位未设置时,元素被清零)。

Intel’s documentation