warp.tile_store#

warp.tile_store(
a: Array[Any],
t: Tile[Any, tuple[int, ...]],
offset: tuple[int, ...],
bounds_check: bool,
) → None#

  • Kernel

  • Differentiable

Store a tile to a global memory array.

This method will cooperatively store a tile to global memory using all threads in the block.

param a:

The destination array in global memory

param t:

The source tile to store data from, must have the same data type and number of dimensions as the destination array

param offset:

Offset in the destination array (optional)

param bounds_check:

Needed for unaligned tiles, but can disable for memory-aligned tiles for faster write times

warp.tile_store(
a: Array[Any],
t: Tile[Any, tuple[int, ...]],
offset: int32,
bounds_check: bool,
) → None

  • Kernel

  • Differentiable