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 function

compute_clc_barrier_counts

def compute_clc_barrier_counts[SCHEDULER_THREADS: Int, TMA_LOAD_THREADS: Int, MMA_THREADS: Int, EPILOGUE_THREADS: Int, CLUSTER_SIZE: Int, cta_group: Int]() -> StaticTuple[Int, Int(4)]

Compute CLC barrier arrival counts.

Parameters:

• ​SCHEDULER_THREADS (Int): Number of scheduler threads.
• ​TMA_LOAD_THREADS (Int): Number of TMA load threads.
• ​MMA_THREADS (Int): Number of MMA threads.
• ​EPILOGUE_THREADS (Int): Number of epilogue threads.
• ​CLUSTER_SIZE (Int): Number of CTAs in the cluster.
• ​cta_group (Int): Number of cooperating CTAs in a group, either 1 or 2.

Returns:

StaticTuple[Int, Int(4)]: StaticTuple of (producer, consumer, throttle_producer, throttle_consumer).