Function _mm512_maskz_reduce_round_ph

Extract the reduced argument of packed half-precision (16-bit) floating-point elements in a by the number of bits specified by imm8, and store the results in dst using zeromask k (elements are zeroed out when the corresponding mask bit is not set).

Rounding is done according to the imm8 parameter, which can be one of:

• _MM_FROUND_TO_NEAREST_INT : round to nearest
• _MM_FROUND_TO_NEG_INF : round down
• _MM_FROUND_TO_POS_INF : round up
• _MM_FROUND_TO_ZERO : truncate
• _MM_FROUND_CUR_DIRECTION : use MXCSR.RC - see _MM_SET_ROUNDING_MODE

Exceptions can be suppressed by passing _MM_FROUND_NO_EXC in the sae parameter.

Intel's documentation