Files
LithosAnanke/docs/v4.0.0/plans/2026-11-18-v4-init.md
T
Claude b13d452e66 v4 init.4th, steps 1 and 2: the ACL policy loaded after POST, then the search order
Captain Bob ruled "ACL at boot", loaded "From init.4th", the include
"Compose[d] in FORTH".
- capsules/v4/init.4th, at block 3980: S" v4:ACL.4th" CAPSULE-INCLUDE,
  then STARSHIPOS.
- capsules/v4/ACL.4th is the policy, git mv'ed from v4/capsule/ACL.fth
  and reflowed into chained 64x16 blocks, its text unchanged.
- CAPSULE-INCLUDE in centre.4th places a capsule and LOADs its lowest
  block. It is composed from CAPSULE-OPEN and CAPSULE-LINE, with no new
  C word.
- The mesh loads v4:init.4th after POST and COLD, its one fixed step
  there.

Test first in test_mesh_boot: the parity line, DOE without naming
STARSHIPOS, and ACL-BOOT pinned. The before-run failed.

After: 113 checks pass and POST is 990/990. All 52 tests, test-32 and
ASan/UBSan pass. The dictionary has 405 words. One amd64 session passes
18/18 with "ACL: active".

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017RzZgrf7V7Zk2iczBXDRbm
2026-11-19 11:57:42 +00:00

6.5 KiB
Raw Blame History

v4's init.4th -- design, for ruling

Written 2026-11-18. Captain Bob ruled: "Write v4's init.4th". That was after the audit of Phase 7 Tasks 5–7 found that v4 has no init.4th.

Nothing here is built. Each point in section 5 is put to Captain Bob one at a time.

1. What the record asks

  • MACHINE.md §3, step 5: "init.4th, the system's init (section 7, item 6). From here the boot is init.4th only. No boot policy is written in C."
  • MACHINE.md §1.1: "the system's init, init.4th, brings it up:
    • it turns on security;
    • it starts the system a module at a time and a node at a time;
    • whoever is logged in gets a prompt: Zuse, rajames, 00 to 09, or whoever it is."
  • MACHINE.md §5: "Everything after POST is init.4th's."
  • Ruled 2026-11-18: init.4th sets the search order, starting in STARSHIPOS, then FORTH. COLD stays as FORTH-79's, with one vocabulary.

2. What the boot does now (v4/system/mesh.c, kernel/src/v4/sk_v4.c)

  1. The ROM assembles the nucleus on Hera.
  2. The kernel words are made.
  3. The capsules are loaded:
    • forth79.4th;
    • the kernel's list in C (sk_v4.c:165): starshipos.4th, centre.4th, hera.4th, physics.4th.
  4. Seal, then POST (990 cases), then COLD.
  5. The kernel's C code runs:
    • the identity gate: no identity drive means headless, and Zuse's drive gives a console (sk_v4_identity.c);
    • the prompt, [zuse@Hera] ok>.

Security is never turned on in the kernel's boot. The policy, v4/capsule/ACL.fth (v3's ACL.4th as FORTH source), is loaded only by a host test (test_host_quit.c:1637). It is not in the kernel's capsule directory, which is built from capsules/ only.

3. The design

capsules/v4/init.4th, a capsule like the others:

  • It is in the kernel's directory, and mkcapsule checks, hashes and signs it.
  • Its blocks are 8300 up, following physics.4th's 8200 range. The range is open for ruling.

The loader runs it last. It is loaded by name after COLD, on Hera, through the same loader as the system capsules. It is the only boot policy, so the loader's one fixed step after POST is "load v4:init.4th". The list before POST stays the loader's (MACHINE.md §5).

What it does, in order, each step its own task, test first:

  1. The search order: STARSHIPOS first, then FORTH (ruled).
  2. Security on:
    • it loads the ACL policy, ACL.fth brought into capsules/v4/ as a capsule;
    • ACL-BOOT runs;
    • BIRTH is pinned as the parent plan's Phase 4 asks (CAPSULE-BIRTH has no v4 counterpart: to be ruled).
  3. The system a module and a node at a time: births through Hera (centre.4th's BIRTH). Which nodes, and in what order, is for ruling. Today the kernel births none at boot.
  4. The prompt for whoever is logged in. Today the identity gate and the prompt are the kernel's C code. Moving them into init.4th is the largest step, and comes last.

4. How each step is verified (named before it is built)

  • test_mesh_boot.c checks each step on the host, seen to fail first:
    • for step 1, S" ok" DOE works at the prompt without naming STARSHIPOS;
    • for step 2, a strict word refuses as test_host_quit.c's ACL cases do.
  • POST stays at 990/990 on all three ISAs. POST runs before init.4th, so it is unchanged.
  • The protocol runs on a clean tree. dict_hash changes with each step's words, and the new value is recorded each time.

5. For ruling, one at a time

  1. Start with step 1 only (the search order), built and accepted, then step 2?
  2. init.4th's blocks: 8300 up?
  3. ACL.fth into capsules/v4/ as a capsule loaded by init.4th, or does init.4th carry the policy itself?
  4. CAPSULE-BIRTH: v4 has no such word. Is it pinned once it exists, or dropped?
  5. Which nodes init.4th births at boot, and in what order.
  6. Step 4: when the identity gate and the prompt move from C into init.4th.

6. Rulings so far (2026-11-18)

  1. Step 1 first (the search order), each step accepted before the next.

  2. Held. Asked where init.4th's blocks go, Captain Bob answered: "It shouldn't matter. Capsules should not occupy fixed ranges." Then: "Capsules are ephemeral. We need to talk later." This runs against the tree as it is:

    • every capsule's Block NNNN headers;
    • mkcapsule's check of them;
    • the ranges in BLOCK_MAP.md and experiments/bare_metal/README.md.

    It may also change what init.4th is: a capsule, or not. Nothing of this design is built until that conversation.

7. Steps 1 and 2 built (2026-11-19)

Ruled: "ACL at boot", first of the new work after v4.0.0-alpha.5. The policy loads "From init.4th", so init.4th comes off hold. The include is to be "Compose[d] in FORTH". Capsules are ephemeral and placed by block (2026-11-18-capsule-blocks.md), which settles the hold.

What is built:

  • capsules/v4/init.4th, at block 3980, so it lands at RAM 1932. It runs S" v4:ACL.4th" CAPSULE-INCLUDE (security on), then STARSHIPOS (the search order). The order is last-first because ACL.4th ends in FORTH DEFINITIONS.
  • capsules/v4/ACL.4th: the policy, git mved from v4/capsule/ACL.fth. It is reflowed into four 64×16 blocks chained with --> (3984–3987, landing at 1936), and its text is unchanged.
  • CAPSULE-INCLUDE ( baddr u -- ) in centre.4th, composed from CAPSULE-OPEN and CAPSULE-LINE: the capsule is placed, and its lowest block is LOADed. It has a dictionary entry.
  • The mesh loads v4:init.4th after POST and COLD, after the PARITY:V4_SYSTEM line, so that line's dict_hash stays the sealed system's. This is the loader's one fixed step there.
  • test_host_quit types ACL.4th as it types any capsule, skipping its Block and --> lines.

Test first (logs/20261119-init-acl/):

  • test_mesh_boot: init.4th's parity line; DOE found without naming STARSHIPOS; ACL-BOOT and ACL-RECHECK pinned. The before-run failed.
  • One false pass was found and corrected on the way. The two checks had first been placed after the test's own COLD, where the search-order check passed only because the line before it named STARSHIPOS. They now run straight after the boot.

Results:

  • 113 checks pass, with POST 990/990.
  • All 52 tests at 64 bits, test-32 and ASan/UBSan pass.
  • The dictionary gate passes with 405 words.
  • One amd64 kernel session passes 18/18, with "ACL: active" at boot.

Still open, one at a time: BIRTH pinned; the countdown "at call"; CAPSULE-BIRTH; births by init.4th; the identity gate and the prompt moving from C into init.4th.