VFNMADD

Operation

Performs fused negative multiply-add on corresponding active floating-point elements.

Assembler Syntax

VFNMADD.{H|S|D}(z) Pn(p), Vn(v), Vn(w), Vn(y)

Privilege

Unprivileged

Required CPUID flags

FP

VECTOR

VECTORFP

Each active H, S, or D lane forms the negated fused multiply-add expression from its two source elements and old accumulator destination element. The complete expression has unbounded intermediate range and precision and is rounded only once when written to the destination. An inactive accumulator lane retains its previous value.

2.75in

Encodings

VFNMADD.{H|S|D}(z) Pn(p), Vn(v), Vn(w), Vn(y)

Format: Instruction format for VFNMADD.{H|S|D}(z) Pn(p), Vn(v), Vn(w), Vn(y)

size — allowed: 0x1..0x3 [reserved_vector_element_type]