IMPORTANT: To view this page as Markdown, append `.md` to the URL (e.g. /get-started.md). For the complete documentation index, see llms.txt.
Skip to main content
For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /get-started.md).

Mojo struct

ClusterLaunchConfig

struct ClusterLaunchConfig

A launch configuration that groups thread blocks into clusters.

Launching with this configuration fails to compile on a target without cluster launch.

Fields​

  • ​grid_dim (Dim): The grid dimensions.
  • ​block_dim (Dim): The block dimensions.
  • ​cluster_dim (Dim): The cluster dimensions.
  • ​shared_mem_bytes (Int): The dynamic shared memory per thread block, in bytes.

Implemented traits​

AnyType, BlockExecutionConfig, ClusterExecutionConfig, Copyable, Deinitable, ExecutionConfig, GridBlockExecutionConfig, ImplicitlyCopyable, Movable, NearComputeGeneralPurposeScratchpadExecutionConfig, RegisterPassable, TrivialRegisterPassable

Methods​

__init__​

def __init__(*, grid_dim: Dim, block_dim: Dim, cluster_dim: Dim, shared_mem_bytes: Int = Int(0)) -> Self

Creates a cluster launch configuration.

Args:

  • ​grid_dim (Dim): The grid dimensions.
  • ​block_dim (Dim): The block dimensions.
  • ​cluster_dim (Dim): The cluster dimensions.
  • ​shared_mem_bytes (Int): The dynamic shared memory per thread block, in bytes.

get_grid_dim​

def get_grid_dim(self) -> Dim

Returns the grid dimensions.

Returns:

Dim: The grid dimensions.

get_block_dim​

def get_block_dim(self) -> Dim

Returns the block dimensions.

Returns:

Dim: The block dimensions.

get_cluster_dim​

def get_cluster_dim(self) -> Dim

Returns the cluster dimensions.

Returns:

Dim: The cluster dimensions.

get_near_compute_scratchpad_usage​

def get_near_compute_scratchpad_usage(self) -> Int

Returns the dynamic shared memory per thread block.

Returns:

Int: The dynamic shared memory per thread block, in bytes.

Was this page helpful?