nx_sota_why_gate.nx
buildroot/runtime/nx_sota_why_gate.nx
about
nx_sota_why_gate.nx -- proves the per-domain board reader, BOTH POLARITIES.
WHAT IS ACTUALLY UNDER TEST HERE IS THE REFUSALS. nx_sota_why has one job people will trust it for --
"where does this domain stand" -- and five ways to be wrong about it that all look like an answer:
answering from a ledger that is absent, truncated, self-inconsistent or stale, or answering about a
DIFFERENT domain whose name merely contains this one. Each gets a tooth with the EXIT CODE pinned,
because the exit code is the contract for every caller that is not a human reading prose.
FIXTURES ARE ASSEMBLED AT RUNTIME under /tmp/<gate>_<clock>/ via gk_fixture_dir, never in
knowledge/store: a gate that shares a directory with a production beat measures the STATE OF THAT
DIRECTORY and reports it as a verdict about the code. The subject is located from /proc/self/exe, not
from the CWD, because the end-to-end teeth deliberately chdir into the fixture tree and a relative elf
path would then resolve to nothing -- which reads as a crashed subject rather than as a moved harness.
license_tier: ORIGINAL expect_exit: 0 No hw writes (Rule 26).
dependencies 5 imports · 0 importers
imports: nx_syscalls.nxnx_gatekit_lib.nxnx_gate_verdict.nxnx_string_ops.nxnx_sota_why_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
| 21 | const SG_PATH: i64 = 512 |
| 22 | const SG_OUT: i64 = 65536 |
| 23 | const SG_BIG: i64 = 1048576 |
| 24 | const SG_SLASH: i64 = 47 |
| 25 | const SG_BIGROWS: i64 = 3200 |
| 26 | const SG_LEDGER_CAP: i64 = 262144 // must equal SW_CAP in the subject; T11 asserts we exceeded it |
| 27 | const SG_ROW1_END: i64 = 33 // end of the first fixture row -- MEASURED by T2, not guessed: my first draft said 34 and the fixture-reached tooth caught it |
functions
| 30 | func sg_selfdir(out: *u8) -> i64 |
| 41 | func main() -> i64 |