pub struct CudaConfig {
pub enabled: bool,
pub device_ordinal: usize,
pub hot_tier_bytes: u64,
pub pinned_host_bytes: u64,
pub transfer_stream_count: usize,
pub layer_streaming: bool,
pub prefer_direct_host_dma: bool,
pub pinned_staging_threshold_bytes: usize,
pub allow_cpu_fallback: bool,
}Expand description
Runtime configuration for the optional CUDA/GPU tier.
The fields are kept in the core config surface so operators can describe a GPU-tier budget and staging policy in one place even when running on a CPU- only build. The direct CUDA runtime remains feature-gated.
Fields§
§enabled: bool§device_ordinal: usize§hot_tier_bytes: u64§pinned_host_bytes: u64§transfer_stream_count: usize§layer_streaming: bool§prefer_direct_host_dma: bool§pinned_staging_threshold_bytes: usize§allow_cpu_fallback: boolTrait Implementations§
Source§impl Clone for CudaConfig
impl Clone for CudaConfig
Source§fn clone(&self) -> CudaConfig
fn clone(&self) -> CudaConfig
Returns a duplicate of the value. Read more
1.0.0 (const: unstable) · Source§fn clone_from(&mut self, source: &Self)
fn clone_from(&mut self, source: &Self)
Performs copy-assignment from
source. Read moreSource§impl Debug for CudaConfig
impl Debug for CudaConfig
Source§impl Default for CudaConfig
impl Default for CudaConfig
Source§impl<'de> Deserialize<'de> for CudaConfigwhere
CudaConfig: Default,
impl<'de> Deserialize<'de> for CudaConfigwhere
CudaConfig: Default,
Source§fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>where
__D: Deserializer<'de>,
fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>where
__D: Deserializer<'de>,
Deserialize this value from the given Serde deserializer. Read more
Source§impl PartialEq for CudaConfig
impl PartialEq for CudaConfig
Source§fn eq(&self, other: &CudaConfig) -> bool
fn eq(&self, other: &CudaConfig) -> bool
Tests for
self and other values to be equal, and is used by ==.Source§impl Serialize for CudaConfig
impl Serialize for CudaConfig
impl Eq for CudaConfig
impl StructuralPartialEq for CudaConfig
Auto Trait Implementations§
impl Freeze for CudaConfig
impl RefUnwindSafe for CudaConfig
impl Send for CudaConfig
impl Sync for CudaConfig
impl Unpin for CudaConfig
impl UnsafeUnpin for CudaConfig
impl UnwindSafe for CudaConfig
Blanket Implementations§
Source§impl<T> BorrowMut<T> for Twhere
T: ?Sized,
impl<T> BorrowMut<T> for Twhere
T: ?Sized,
Source§fn borrow_mut(&mut self) -> &mut T
fn borrow_mut(&mut self) -> &mut T
Mutably borrows from an owned value. Read more