v_fract_f64 GPU Native ISA AMD Vector
V FRACT F64 Vector Arithmetic
Compute the fractional portion of a double-precision float input and store the result in floating point format into a vector register.
Encoding
Operands
Operand details have not yet been curated for this instruction.
GFX Target Compatibility
| Target | Support |
|---|---|
| gfx1100 | ✅ Supported |
0.5ULP accuracy, denormals are accepted. This is intended to comply with the DX specification of fract where the function behaves like an extension of integer modulus; be aware this may differ from how fract() is defined in other domains. For example: fract(- 1.2) = 0.8 in DX. Obey round mode, result clamped to 0x3fefffffffffffff.
Related
More in Vector Arithmetic
Reference
Semantics
Example
v_fract_f64 v[5:6], -1A real instruction accepted by the LLVM assembler, taken verbatim from LLVM's own AMDGPU MC test suite (gfx11). Not from an AMD document, and not authored here.
Sources
- User Guide for AMDGPU Backend ↗ - LLVM Project
-
"AMD Instinct MI300" Instruction Set Architecture: Reference Guide ↗
- Advanced Micro Devices, Inc.
Reference Guide, page 202. - LLVM MC assembler tests for AMDGPU (gfx11) ↗ - LLVM Project