code wiki / _hdl_build / nx_ad_core.nx

nx_ad_core.nx

buildroot/runtime/_hdl_build/nx_ad_core.nx

7697 B174 linesdepth 3pulls 3 transitivereach 4 importersview sourcekind librarytopic ad
docsdependenciesstructsconstsfunctions

about

nx_ad_core.nx -- AUTORUN BEAT CORE (mainless; X-Q-007 "beat list as DATA"). The 6h slow beat's TARGETS come from conf rows, not code: T <basename> one durable-lane target per row (order preserved, cap 8) No T rows -> the historical default (nx_math_rung_run) so the existing conf keeps working unchanged. adc_beat runs each target sequentially via _offc/nx_sov_build_run.elf (rebuild-from-source = the evidence renewal), appends one AUTORUN-TGT row per target and ONE AUTORUN-BEAT row keeping the historical rung_rc= field (= worst rc) plus targets=<n>. license_tier: ORIGINAL

dependencies 2 imports · 4 importers

nx_syscalls.nx nx_itoa_lib.nx nx_ad_core.nx _ad_dupguard_test.nx _ad_selfmeasure_test.nx _autorun_targets_gate_authored.nx nx_autorun_daemon.nx

imports: nx_syscalls.nxnx_itoa_lib.nx

imported by: _ad_dupguard_test.nx_ad_selfmeasure_test.nx_autorun_targets_gate_authored.nxnx_autorun_daemon.nx

structs

none

consts

13const ADC_MAGIC_8192: i64 = 8192
15const ADC_MAXTGT: i64 = 8
16const ADC_NAMECAP: i64 = 64

functions

31func ad_acquire_lock(path: *u8) -> i64
called by 2: mainmain
37func adc_w(fd: i64, s: *u8) -> i64 { var n: i64=0; while s[n]!=(0 as u8){n=n+1} sys_write(fd,s,n); return 0 }
called by 1: adc_beat calls 1: sys_write
42func adc_wn(fd: i64, v: i64) -> i64 { nxi_fd(fd, v); return 0 }
called by 2: mainadc_beat calls 1: nxi_fd
44func adc_read(path: *u8, buf: *u8, cap: i64) -> i64
56func adc_targets(confpath: *u8, arena: *u8) -> i64
called by 2: mainadc_beat calls 2: sys_mmapadc_read
108func adc_run_one(name: *u8, outpath: *u8) -> i64
132func adc_beat(confpath: *u8, outpath: *u8, logpath: *u8) -> i64