pub const BA_GATES_BLOCK: u32 = 256;Expand description
Threads per CTA — the parent’s block, and a contract: the reduction order
is a function of this, BA_GATES_LANES and BA_GATES_OUTS.
pub const BA_GATES_BLOCK: u32 = 256;Threads per CTA — the parent’s block, and a contract: the reduction order
is a function of this, BA_GATES_LANES and BA_GATES_OUTS.