rsqrt GPU Virtual ISA NVIDIA

Reciprocal Square Root (Approximate) Arithmetic

rsqrt.approx.f32 d, a;

Fast hardware approximation of 1/sqrt(x).

Encoding

PTX is a virtual instruction set. It has no single, stable native binary encoding - the compiler lowers this instruction to different native machine code depending on the selected NVIDIA target architecture (compute capability). This page intentionally shows no bit-diagram; see the target/version requirements below for what governs how this instruction compiles.
PTX ISA Version Introduced PTX ISA 1.0
Minimum Target sm_10

Syntax Forms

One mnemonic covers many type / state-space / scope / modifier combinations - each row below is an independently valid form.

Syntax Data Types State Space(s) Modifiers Min. Target Description
rsqrt.approx.f32 d, a; f32 approx sm_10 Reduced-precision reciprocal square root.

Operands

  • d
    Destination register
  • a
    Source operand

At a Glance

Data Types f32
Modifiers approx

Related AMDGPU Concepts

Special-Function Approximation ↗
equivalent with restrictions
v_rsq_f32 (AMDGPU)
v_sin_f32 (AMDGPU)
v_sqrt_f32 (AMDGPU)
v_rcp_f32 (AMDGPU)

Related

More in Arithmetic

Reference

NVIDIA PTX ISA

Description

Compute 1/sqrt(a) and store the result in d.

Semantics

d ≈ 1 / sqrt(a), with a hardware-specific ULP error bound rather than a fully IEEE-rounded result.

Examples

rsqrt.approx.ftz.f32  isr, x;
rsqrt.approx.f64      ISR, X;

Reproduced from NVIDIA's official PTX ISA documentation for technical accuracy.

Sources