code wiki / (root) / nx_pardemo_heavy.nx

nx_pardemo_heavy.nx

buildroot/runtime/nx_pardemo_heavy.nx

659 B23 linesdepth 8pulls 10 transitivereach 0 importersview sourcekind tool
docsdependenciesstructsconstsfunctions

about

compute-heavy parallel_for via the (now-buildable) thread_pool: 4 workers, 32 indices, each a 2e6 LCG loop. Probe should read ~4 cores. Proves the ERGONOMIC pool API delivers real multi-core.

dependencies 1 imports · 0 importers

nx_parallel.nx nx_pardemo_heavy.nx

imports: nx_parallel.nx

imported by: nobody (leaf or entry point)

call flow from main pre-order; caps 40 nodes / depth 6 declared; ↻ = already shown

main nx_parallel_for sys_mmap nx_atom_load_i64 _nx_chunk_start _nx_chunk_end nx_pool_submit nx_atom_faa_i64 nx_chan_send nx_chan_try_send _nx_chan_cell nx_thread_yield _pool_futex_wake_all sys_futex_wake nx_pool_shutdown nx_atom_faa_i64 ↻ nx_chan_send ↻ _pool_futex_wake_all ↻ nx_atom_load_i64 ↻ nx_thread_yield ↻

structs

none

consts

none

functions

7func heavy_body(i: i64) -> i64
15func main() -> i64