code wiki / _hdl_build / nx_pub_submit_gate.nx

nx_pub_submit_gate.nx

buildroot/runtime/_hdl_build/nx_pub_submit_gate.nx

5269 B105 linesdepth 6pulls 11 transitivereach 0 importersview sourcekind gate/prooftopic pub
docsdependenciesstructsconstsfunctions

about

nx_pub_submit_gate.nx -- R1 proof for THE NISHI PUBLISHER intake: under a REAL multi-process race (PG_N sys_fork children x PG_K pub_submit calls each = the many-workstreams-publish-at-once scenario), the durable framed queue keeps EVERY request and tears NONE -- while the UNLOCKED neg-control (same record, multi-write, no flock) DOES tear under identical load. That is the exact collision that crashed sites.elf, measured before/after. Sovereign: our fork/wait4 + our framed-append + our scan. Per-wsid test queue (nx_runpath) so sibling workstreams' gate runs don't contend. GREEN iff framed = all-present & zero-torn AND unlocked tears. license_tier: ORIGINAL expect_exit: 0

dependencies 3 imports · 0 importers

nx_syscalls.nx nx_publisher.nx nx_runpath.nx nx_pub_submit_gate.nx

imports: nx_syscalls.nxnx_publisher.nxnx_runpath.nx

imported by: nobody (leaf or entry point)

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

main sys_mmap g_w sys_write rp_wsid sys_openat_rd sys_mmap ↻ sys_read sys_close rp_match rp_ensure sys_mmap ↻ rp_path pg_make_src sys_openat_wr sys_write ↻ sys_close ↻ pg_phase pg_reset sys_openat_wr ↻ sys_close ↻ sys_fork pg_child sys_mmap ↻ pg_dest fa_cat fa_catn sys_mmap ↻ pub_submit_to sys_mmap ↻ pub_sha_file sys_mmap ↻ sys_read_file sys_openat_rd ↻ sys_lseek sys_mmap ↻ sys_read ↻ sys_close ↻ sha256_digest sys_mmap ↻

structs

none

consts

12const PG_N: i64 = 4
13const PG_K: i64 = 100

functions

15func g_w(s: *u8) -> i64 { var n: i64=0; while s[n]!=(0 as u8){n=n+1} sys_write(1,s,n); return 0 }
called by 3: g_ng_rowmain calls 1: sys_write
16func g_n(v: i64) -> i64 { var m: i64=v; if m<0{g_w("-" as *u8);m=0-m} let t:*u8=sys_mmap(24); var k:i64=0; if m==0{t[0]=48 as u8;k=1} while m>0{t[k]=(48+(m%10)) as u8;m=m/10;k=k+1} var i:i64=0; let o:*u8=sys_mmap(24); while i<k{o[i]=t[k-1-i];i=i+1} sys_write(1,o,k); return 0 }
called by 1: main calls 3: g_wsys_mmapsys_write
17func g_row(id: *u8, ok: i64, pass: *i64) -> i64 { g_w(" " as *u8); g_w(id); g_w(": " as *u8); if ok==1 { g_w("OK\n" as *u8); pass[0]=pass[0]+1 } else { g_w("FAIL\n" as *u8) } return 0 }
called by 1: main calls 1: g_w
20func pg_dest(child: i64, k: i64, out: *u8) -> i64
called by 1: pg_child calls 2: fa_catfa_catn
26func pg_make_src(srcpath: *u8) -> i64
called by 1: main calls 3: sys_openat_wrsys_writesys_close
33func pg_reset(qpath: *u8) -> i64 { let fd: i64 = sys_openat_wr(qpath, 0x1a4); if fd >= 0 { sys_close(fd) } return 0 }
called by 1: pg_phase calls 2: sys_openat_wrsys_close
36func pg_scan(qpath: *u8, out: *i64) -> i64
called by 1: pg_phase calls 2: sys_mmapsys_read_file
56func pg_child(qpath: *u8, srcpath: *u8, child: i64, good: i64) -> i64
called by 1: pg_phase calls 5: sys_mmappg_destpub_submit_topub_append_unlockedsys_exit
68func pg_phase(qpath: *u8, srcpath: *u8, good: i64, out: *i64) -> i64
called by 1: main calls 6: pg_resetsys_forkpg_childsys_mmapsys_wait4pg_scan
79func main() -> i64
calls 10: sys_mmapg_wrp_wsidrp_ensurerp_pathpg_make_src+4