nx_provcensus_gate.nx
buildroot/runtime/nx_provcensus_gate.nx
about
nx_provcensus_gate.nx -- proves the fleet provenance-drift census's judgement half (2026-08-26).
The subject is nx_provcensus_lib.nx, imported so it compiles INTO this binary. That is the whole
reason the lib exists as a separate file: this gate can be mutation-proven, whereas a gate that
fork/execs the deployed nx_provcensus.elf could not be -- every mutant would read NOT-REACHED and the
harness would report a fault in the gate rather than in the subject.
What is actually at stake. The census exists to answer "how many ships proved the wrong subject", and
the only way it can lie in the FLATTERING direction is by letting something that is not a measured
CURRENT read as CURRENT. So the teeth below spend most of their effort on the three ways to fail to
observe -- an empty capture, a capture with no verdict token, and a verdict token this build has never
heard of -- and require all three to land in UNREADABLE.
license_tier: ORIGINAL No hw writes (Rule 26). expect_exit: 0
dependencies 2 imports · 0 importers
imports: nx_gate_verdict.nxnx_provcensus_lib.nx
imported by: nobody (leaf or entry point)
call flow from main pre-order; caps 40 nodes / depth 6 declared; ↻ = already shown
structs
| none |
consts
| 17 | const PG_COUNTS_BYTES: i64 = 40 // PCX_NBUCKET slots of 8; the lib owns the bucket count |
| 20 | const PG_CRBUF: i64 = 128 |
| 21 | const PG_CR: i64 = 13 |
| 22 | const PG_LF: i64 = 10 |
functions
| 27 | func pg_put(d: *u8, o: i64, s: *u8) -> i64 called by 1: main |
| 40 | func pg_streq(a: *u8, b: *u8) -> i64 called by 1: main |
| 50 | func main() -> i64 |