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

VRSQRT28SS

Computes the approximate reciprocal square root of a scalar single-precision floating-point value. The result is the reciprocal square root of the source operand with a relative error of less than $2^{-12}$.

The table after the description covers what the source and destinations can be.

DO NOT support LOCK

This instruction is available only in 64-bit mode. It requires the SSE4.1 instruction set extension.

The precision of the result is limited; it provides an approximation and SHOULD NOT be used for high-precision calculations without subsequent Newton-Raphson iterations. If the source operand is $\pm0$, the result is $\pm\infty$ and #Z is raised. If the source operand is $\text{NaN}$, the result is a $\text{qNaN}$ and #I is raised. If the source operand is negative, the result is $\text{qNaN}$ and #I is raised. The instruction MAY trigger #P if the result is inexact.