Struct SimpleConv 

pub struct SimpleConv<TMM: TileMatmulFamily, LL: FullLoadingStrategy<RuntimeArgs>, LR: FullLoadingStrategy<RuntimeArgs>> { /* private fields */ }

Cmma convolution

Trait Implementations

impl<TMM: TileMatmulFamily<LhsTile = Strided, RhsTile = Strided, AccTile = Option<Strided>, OutTile = Strided>, LL: FullLoadingStrategy<RuntimeArgs, TileKind = Strided>, LR: FullLoadingStrategy<RuntimeArgs, TileKind = Strided, SyncStrategy = LL::SyncStrategy>> Algorithm for SimpleConv<TMM, LL, LR>

type Routine = SimpleAlgorithm<TMM, LL, LR, SyncBiasLoading>

type Args = TensorArgs<RuntimeArgs>

fn correct_layout<R: Runtime>( client: &ComputeClient<R>, handle: TensorBinding<R>, dtype: StorageType, _operation: ConvolutionOperation, ) -> Result<TensorBinding<R>, LaunchError>

const IS_SPECIALIZED: bool = false

Whether to select specialized load flow in tests. Should replace with something cleaner eventually, but this is nice and simple.

fn filter_vector_sizes( vector_sizes: AvailableVectorSizes, ) -> AvailableVectorSizes