Accesses an element by coordinate

Returns the stride of the layout

Before entry of the subtile load loop

Elect one thread in the warp. The elected thread gets its predicate set to true, all others obtain false.

Set the imaginary part of the complex number

Set the real part of the complex number

Return TmaDescriptor/TensorMap

Unpacks an element from memory

CUTE helper to cast SMEM pointer to unsigned

< Efficiently disables all accesses guarded by mask

Returns the ceiling of (a / b)

Returns the layout object

Returns a pointer to the managed object or nullptr if no object is owned.

Accesses the real part of the quaternion

Obtains a Coord<2> from GemmCoord

Overrides the internal iteration index

Prevents early exit of producer blocks in Cluster. This should be called once before kernel exits.

Converts to float

Barrier arrive on local smem

< Efficiently disables all accesses guarded by mask

Obtains the threadblock offset (in units of threadblock-scoped tiles)

Computes the offset of an index from the origin of the tensor

Returns a warp-uniform value indicating the canonical warp index of the calling threads. Threads within the warp must be converged.