Struct FlashArmCapability 

pub struct FlashArmCapability {
    pub cuda_flash_kernel: bool,
    pub cuda_in_topology: bool,
}

The capability + topology gate for the CUDA flash decode arm: a CUDA FlashAttn kernel must be bound and a CUDA device must be present in the current topology. Split out (and injectable) so the gate logic is unit-testable without a CUDA build, while production reads the real process-global registry + topology via Self::production.

Fields

cuda_flash_kernel: bool

A (FlashAttn, Cuda) kernel is bound (static or runtime-adopted).

cuda_in_topology: bool

A CUDA device is present in the current SystemTopology.

Implementations

impl FlashArmCapability

pub fn production() -> Self

Read the real process-global capability.

pub fn available(&self) -> bool

The flash arm may be offered only when both halves hold.

Trait Implementations

impl Clone for FlashArmCapability

fn clone(&self) -> Self

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Copy for FlashArmCapability

impl Debug for FlashArmCapability

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

Formats the value using the given formatter.