Enum Feature 

#[non_exhaustive]
pub enum Feature {
    StreamOrderedAllocator,
    VirtualMemoryManagement,
    LibraryManagement,
    GreenContexts,
    MulticastObjects,
    GraphConditionalNodes,
    GraphSwitchNodes,
    CudaLaunchKernelEx,
    NvJitLink,
    TensorMapObjects,
    RuntimeLogBuffer,
    CudaInitDevice,
    RuntimeGreenContexts,
}

Named CUDA capabilities, each gated by the minimum toolkit/driver version in which they became available. Safe-API crates call supports before invoking the underlying symbol and surface Error::FeatureNotSupported when the installed driver is too old.

Variants (Non-exhaustive)

Non-exhaustive enums could have additional variants added in future. Therefore, when matching against variants of non-exhaustive enums, an extra wildcard arm must be added to account for any future variants.

StreamOrderedAllocator

Stream-ordered memory allocator (cuMemAllocAsync, pools). CUDA 11.2+.

VirtualMemoryManagement

Virtual memory management (cuMemCreate, cuMemMap, …). CUDA 10.2+.

LibraryManagement

Context-independent library loading (cuLibraryLoadData). CUDA 12.0+.

GreenContexts

Lightweight SM-partitioned contexts. CUDA 12.0+.

MulticastObjects

Multicast objects across multiple GPUs. CUDA 12.0+.

GraphConditionalNodes

Graph conditional (IF/WHILE) nodes. CUDA 12.3+.

GraphSwitchNodes

Graph SWITCH nodes. CUDA 12.8+.

CudaLaunchKernelEx

Extensible kernel launch (cuLaunchKernelEx, cluster dims). CUDA 12.0+.

NvJitLink

nvJitLink modern JIT linker. CUDA 12.0+.

TensorMapObjects

Hopper-only TMA tensor map descriptors. CUDA 11.8+ (HW 9.0+).

RuntimeLogBuffer

Runtime-API log buffer (cudaLogs*). CUDA 12.0+.

CudaInitDevice

cudaInitDevice (initialize primary context without making it current). CUDA 12.0+.

RuntimeGreenContexts

Green-context-aware execution context APIs on the runtime. CUDA 13.1+.

Implementations

impl Feature

pub const fn required_version(self) -> CudaVersion

The minimum CUDA version at which the feature is available.

Trait Implementations

impl Clone for Feature

fn clone(&self) -> Feature

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for Feature

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

Formats the value using the given formatter.

impl Hash for Feature

fn hash<__H: Hasher>(&self, state: &mut __H)

Feeds this value into the given Hasher.

fn hash_slice<H>(data: &[Self], state: &mut H)

where H: Hasher, Self: Sized,

Feeds a slice of this type into the given Hasher.

impl PartialEq for Feature

fn eq(&self, other: &Feature) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Copy for Feature

impl Eq for Feature

impl StructuralPartialEq for Feature