LinearLayout

Enum LinearLayout 

Source
pub enum LinearLayout {
    Plain(PlainLayout),
    Strided(StridedLayout),
    Permuted(PermutedLayout),
}
Expand description

Maps a linear index based on line count to a potentially strided tensor. Only applies the necessary level of striding, either none, only the last dim (for freshly allocated strided tensors), or all dimensions.

Treats indices as the line index, with the shape being adjusted for line size.

Layout version of [index_offset_contiguous]

Variants§

§

Plain(PlainLayout)

Input is contiguous, no mapping

§

Strided(StridedLayout)

Strided tensor, i.e. freshly allocated but not permuted

§

Permuted(PermutedLayout)

Permuted layout, tracks the entire shape/strides and not just the last dim

Implementations§

Trait Implementations§

Source§

impl Clone for LinearLayout

Source§

fn clone(&self) -> LinearLayout

Returns a duplicate of the value. Read more
1.0.0 · Source§

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

Performs copy-assignment from source. Read more
Source§

impl CubeDebug for LinearLayout

Source§

fn set_debug_name(&self, scope: &mut Scope, name: &'static str)

Set the debug name of this type’s expansion. Should do nothing for types that don’t appear at runtime
Source§

impl CubeType for LinearLayout

Source§

type ExpandType = LinearLayoutExpand

Source§

fn into_mut(scope: &mut Scope, expand: Self::ExpandType) -> Self::ExpandType

Wrapper around the init method, necessary to type inference.
Source§

impl LaunchArg for LinearLayout

Source§

type RuntimeArg<'a, R: Runtime> = LinearLayoutArgs<'a, R>

The runtime argument for the kernel.
Source§

type CompilationArg = LinearLayoutCompilationArg

Compilation argument.
Source§

fn compilation_arg<'a, R: Runtime>( runtime_arg: &Self::RuntimeArg<'a, R>, ) -> Self::CompilationArg

Source§

fn expand( arg: &Self::CompilationArg, builder: &mut KernelBuilder, ) -> <Self as CubeType>::ExpandType

Register an input variable during compilation that fill the KernelBuilder.
Source§

fn expand_output( arg: &Self::CompilationArg, builder: &mut KernelBuilder, ) -> <Self as CubeType>::ExpandType

Register an output variable during compilation that fill the KernelBuilder.
Source§

impl Layout for LinearLayout

Source§

type Coordinates = u32

The coordinate type used by the conceptual tensor represented by this layout, i.e. (u32, u32, u32) for a fixed-rank 3D tensor. This does not have to match the rank of the underlying storage (if applicable). It’s only how the tensor is interpreted (viewed) by the code.
Source§

type SourceCoordinates = u32

The coordinate type used by the inner storage wrapped in this layout, i.e. u32 for Array, or (u32, u32) for a 2D view.
Source§

fn to_source_pos(&self, pos: Self::Coordinates) -> u32

Transform a set of n-dimensional coordinates to a source coordinate space. It is recommended to use absolute positions here, and handle the translation into lines at the lowest level (global memory layout).
Source§

fn to_source_pos_checked(&self, pos: Self::Coordinates) -> (u32, bool)

Transform a set of n-dimensional coordinates to an offset into the underlying storage, and return whether the position is in bounds of this layout. See also Layout::to_source_pos
Source§

fn shape(&self) -> Self::Coordinates

The shape of the conceptual tensor represented by this layout. Not necessarily the extent of the underlying storage, but only this view of it.
Source§

fn is_in_bounds(&self, pos: Self::Coordinates) -> bool

Source§

fn __expand_to_source_pos( scope: &mut Scope, this: <Self as CubeType>::ExpandType, pos: <Self::Coordinates as CubeType>::ExpandType, ) -> <Self::SourceCoordinates as CubeType>::ExpandType

Source§

fn __expand_to_source_pos_checked( scope: &mut Scope, this: <Self as CubeType>::ExpandType, pos: <Self::Coordinates as CubeType>::ExpandType, ) -> <(Self::SourceCoordinates, bool) as CubeType>::ExpandType

Source§

fn __expand_shape( scope: &mut Scope, this: <Self as CubeType>::ExpandType, ) -> <Self::Coordinates as CubeType>::ExpandType

Source§

fn __expand_is_in_bounds( scope: &mut Scope, this: <Self as CubeType>::ExpandType, pos: <Self::Coordinates as CubeType>::ExpandType, ) -> <bool as CubeType>::ExpandType

Auto Trait Implementations§

Blanket Implementations§

Source§

impl<T> Any for T
where T: 'static + ?Sized,

Source§

fn type_id(&self) -> TypeId

Gets the TypeId of self. Read more
Source§

impl<T> Borrow<T> for T
where T: ?Sized,

Source§

fn borrow(&self) -> &T

Immutably borrows from an owned value. Read more
Source§

impl<T> BorrowMut<T> for T
where T: ?Sized,

Source§

fn borrow_mut(&mut self) -> &mut T

Mutably borrows from an owned value. Read more
Source§

impl<T> CloneToUninit for T
where T: Clone,

Source§

unsafe fn clone_to_uninit(&self, dest: *mut u8)

🔬This is a nightly-only experimental API. (clone_to_uninit)
Performs copy-assignment from self to dest. Read more
Source§

impl<T> From<T> for T

Source§

fn from(t: T) -> T

Returns the argument unchanged.

Source§

impl<T, U> Into<U> for T
where U: From<T>,

Source§

fn into(self) -> U

Calls U::from(self).

That is, this conversion is whatever the implementation of From<T> for U chooses to do.

Source§

impl<L> LaunchLayout for L

Source§

impl<T> ToOwned for T
where T: Clone,

Source§

type Owned = T

The resulting type after obtaining ownership.
Source§

fn to_owned(&self) -> T

Creates owned data from borrowed data, usually by cloning. Read more
Source§

fn clone_into(&self, target: &mut T)

Uses borrowed data to replace owned data, usually by cloning. Read more
Source§

impl<T, U> TryFrom<U> for T
where U: Into<T>,

Source§

type Error = Infallible

The type returned in the event of a conversion error.
Source§

fn try_from(value: U) -> Result<T, <T as TryFrom<U>>::Error>

Performs the conversion.
Source§

impl<T, U> TryInto<U> for T
where U: TryFrom<T>,

Source§

type Error = <U as TryFrom<T>>::Error

The type returned in the event of a conversion error.
Source§

fn try_into(self) -> Result<U, <U as TryFrom<T>>::Error>

Performs the conversion.
Source§

impl<V, T> VZip<V> for T
where V: MultiLane<T>,

Source§

fn vzip(self) -> V