Mojo function
tiled_matmul_run
tiled_matmul_run[config: KernelConfig, transpose_b: Bool, b_packed: Bool, simd_size: Int, elementwise_epilogue_enabled: Bool, kernel_id: InnerKernelID, algorithm: InnerMatmulKernel](alg: algorithm, c: NDBuffer[type, 2, origin, shape], a: NDBuffer[type, 2, origin, shape], b: NDBuffer[type, 2, origin, shape], elementwise_epilogue_fn: fn(GemmShape, GemmShape) escaping -> None, global_tile_shape: GemmShape, global_tile_offset: GemmShape)
Interface function to run tiled matmul on a given sub-tile.
Args:
- alg (
algorithm
): InnerMatmulKernel algorithm for microkernel. - c (
NDBuffer[type, 2, origin, shape]
): Pre-allocated buffer space for result. - a (
NDBuffer[type, 2, origin, shape]
): Operand A of the matmul. - b (
NDBuffer[type, 2, origin, shape]
): Operand B of the mamtul. - elementwise_epilogue_fn (
fn(GemmShape, GemmShape) escaping -> None
): The elementwise epilogue function. - global_tile_shape (
GemmShape
): Tile shape this call will process. - global_tile_offset (
GemmShape
): Tile offset on the original buffer.
Was this page helpful?
Thank you! We'll create more content like this.
Thank you for helping us improve!