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

pjoin

def pjoin[State: ReduceOp, params: ContextParams, //](mut state: State, mut ctx: Context[params])

Cross-thread join for a single monoid state.

GPU: cooperative warp/block reduce (warp / block tiers) or cross-block atomic finish (split-K); no-op for the tiled and serial tiers. CPU: drives the monoid's own pjoin body via a single-participant SerialReducer — no cross-thread combine, but monoids may run per-state finalization here (e.g. ReduceSum horizontally reducing its SIMD accumulator into the scalar acc field bodies read).

Parameters:

• ​State (ReduceOp): The monoid type being joined.
• ​params (ContextParams): Comptime dispatch parameters.

Args:

• ​state (State): The body's monoid accumulator.
• ​ctx (Context[params]): The dispatch bundle.