Struct Operand 

pub struct Operand {
    pub register: Register,
    pub selection: Selection,
    pub components: u8,
    pub indices: Vec<Index>,
    pub immediate: [u32; 4],
    pub modifier: u8,
    pub precision: u8,
    pub non_uniform: bool,
}

An instruction’s or declaration’s operand.

Fields

register: Register

The kind of register it names.

selection: Selection

Which components it selects.

components: u8

The number of components the operand token states (0, 1 or 4): a one-component operand (oDepth, l(1.0)) and a four-component register selecting one lane (r0.x) both have a Selection::Scalar.

indices: Vec<Index>

Its index dimensions, outermost first: none for an immediate, one for r3, two for cb2[5].

immediate: [u32; 4]

The values of an immediate operand (one lane repeated, or four); zero for other registers. A 64-bit immediate’s four words are two doubles, low word first; one of one component is a single word repeated, as Microsoft’s disassembler reads it (fxc writes none).

modifier: u8

0 none, 1 negate, 2 absolute, 3 negated absolute (float sources); 4 two’s-complement negate (an integer instruction’s negate; the encoding has no such value, so a caller that knows the instruction reads integers sets it).

precision: u8

The minimum-precision annotation of the operand’s extended token (0 none; 1 float16, 2 float2_8, 4 sint16, 5 uint16).

non_uniform: bool

The operand’s extended token marks it as a non-uniform resource index.

Trait Implementations

impl Clone for Operand

fn clone(&self) -> Self

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for Operand

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.