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

VSQRTSH

Computes the square root of a packed signed short floating-point value.

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

DO NOT support LOCK

This instruction is available only when using the AVX extension. It operates on packed 16-bit floating-point values (half-precision).

If the source operand contains a negative value, the result is a NaN (Not-a-Number) and the #I exception is signaled. If the source is zero, the result is zero and no exception is signaled. If the source is a NaN, the result is a NaN.

Precision control: The instruction is subject to the rounding control bits in the MXCSR register. If the result is not exact, the #P exception is signaled.