code wiki / (root) / nx_distexec_gate.nx

nx_distexec_gate.nx

buildroot/runtime/nx_distexec_gate.nx

5661 B130 linesdepth 4pulls 5 transitivereach 0 importersview sourcekind gate/proof
docsdependenciesstructsconstsfunctions

about

nx_distexec_gate.nx -- the correctness a distributed scheduler stands on: the merged result is IDENTICAL for any partition count (partition-invariance), for sum/count/min/max AND group-by-sum, including a partition count that does NOT divide the row count (ragged partitions). If any of these failed, running partitions on different workers would silently return a wrong answer. Hand-computed known answers + a cross-P invariance sweep on a larger frame. D001 verdict via nx_gate_verdict. expect_exit: 0

dependencies 2 imports · 0 importers

nx_gate_verdict.nx nx_distexec.nx nx_distexec_gate.nx

imports: nx_gate_verdict.nxnx_distexec.nx

imported by: nobody (leaf or entry point)

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

main gv_ctr sys_mmap gv_head gv_puts sys_write dg_frame1 cf_encoded_bytes cf_slen cf_align8 cf_encode cf_align8 ↻ de_run_agg cf_nrows sys_mmap ↻ de_bound de_agg_range cf_col cf_data_off de_merge gv_check gv_puts ↻ dg_frame2 cf_encoded_bytes ↻ cf_encode ↻ de_run_groupby cf_nrows ↻ sys_mmap ↻ de_bound ↻ de_gb_range cf_col ↻ dg_eq gv_verdict gv_puts ↻ gv_num sys_mmap ↻ sys_write ↻ sys_munmap gv_journal sys_openat_append

structs

none

consts

none

functions

9func dg_eq(a: i64, b: i64) -> i64 { if a == b { return 1 } return 0 }
called by 1: main
10func dg_frame1(a: *i64, nm: *u8, n: i64) -> *u8
called by 1: main calls 2: cf_encoded_bytescf_encode
20func dg_frame2(a: *i64, b: *i64, na: *u8, nb: *u8, n: i64) -> *u8
called by 1: main calls 2: cf_encoded_bytescf_encode
33func main() -> i64