This content has not been validated; it was AI-generated following the x86-64 ISA Documentation generation process to maintain high quality. However, as no human has verified this data, errors may exist. If the accuracy of this data is critical, please consult the Intel SDM. The model used to generate this is gemma4:31b

VRCP28SS

Computes an approximate reciprocal of a single-precision floating-point value. The instruction calculates $2^{28} / \text{source}$ and stores the result in the destination.

The table below covers what the source and destinations can be.

DO NOT support LOCK

This instruction is available only when the processor is operating in 64-bit mode or compatibility mode. It requires the VEX encoding prefix; without it, the instruction is invalid.

The result is an approximation and SHALL NOT be used for high-precision requirements without subsequent refinement using the Newton-Raphson method (typically via VRSQRTSS or VRCPSS iterations). The precision of the approximation is limited to a relative error of approximately $2^{-12}$. If the input is a signaling NaN, it SHALL trigger #I; if the input is infinity or zero, the result is $0.0$.