code wiki / _hdl_build / nx_synced_push.nx
nx_synced_push.nx
buildroot/runtime/_hdl_build/nx_synced_push.nx
about
nx_synced_push.nx -- S-CLASS LIVE-SYNC: the ARBITER-WRAPPED NAS pusher. Every push to the ONE shared NAS
doc-root goes through a single kernel-enforced "nas_push" lease (nx_arbiter), so concurrent sibling
workstreams SERIALIZE on the vault->ssh chain instead of racing it -- the direct fix for the intermittent
"live-sync bullshit" (two siblings pushing at once corrupted each other's secret/staging). Thin composition
of two PROVEN pieces: nx_arbiter (the lease, race-gate-proven 380->0 lost updates) + the EXACT fork/exec
idiom nx_aw_push itself uses to spawn its vault children (p_run). Same argv contract as nx_aw_push:
argv[1] = src-spec file (one line: the local payload path) argv[2] = dst-cmd file (the remote write cmd)
CWD must be nxc2 (nx_aw_push forks _offc/*.elf by RELATIVE path). license_tier: ORIGINAL
dependencies 3 imports · 0 importers
imports: nx_syscalls.nxnx_itoa_lib.nxnx_arbiter.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
| 13 | const SP_PUSH_ELF: *u8 = "_offc/nx_aw_push.elf" |
| 14 | const SP_RESOURCE: *u8 = "nas_push" |
functions
| 19 | func sp_push(src_spec: *u8, dst_cmd: *u8) -> i64 |
| 40 | func sp_w(s: *u8) -> i64 { var n: i64=0; while s[n]!=(0 as u8){n=n+1} sys_write(1,s,n); return 0 } |
| 45 | func sp_n(v: i64) -> i64 { nxi_out(v); return 0 } |
| 47 | func main(argc: i64, argv: *i64) -> i64 |