-
__m512 amd_vrs16_roundf(__m512 x)#
Computes round per-lane for sixteen single-precision elements.
- Parameters:
x – Input vector x.
- Returns:
Vector of rounded value.
-
__m512d amd_vrd8_round(__m512d x)#
Computes round per-lane for eight double-precision elements.
- Parameters:
x – Input vector x.
- Returns:
Vector of rounded value.