Artemis bus-agnostic discovery: signature format + idle-loop generalization (FABRIC-3.md §XXVI follow-on)
Build / build-amd64-iso (push) Waiting to run
Build / build-aarch64-iso (push) Waiting to run
Build / build-riscv64-img (push) Waiting to run

Step 1: new artemis_sig_t header format (magic 'ARTM', sibling to
homeblocks_sig_t, distinct so a generic scan can tell Artemis's own disk
apart from an identity thumbdrive by content alone) -- artemis_sig.h/.c,
wired into Makefile.starkernel.

Step 2: sk_repl_idle()'s existing per-USB-MSC-slot attach handling (the
pattern WIREBIND already uses for identity thumbdrives) now also checks
for the ARTM signature whenever a device's home-blocks check comes back
BLANK. On a match, once Artemis's own storage-attach round-trip
(HERA-BLK-ATTACH-REQ/BLK-ATTACH-ACK) confirms success,
capsule_zuse_boot_load_root_pubkey() runs -- the same call kernel_main.c's
synchronous QEMU-only virtio-blk path already makes, now reachable
without a hardcoded PCI vendor/device scan. That function is already
idempotent (no-op once zuse_root_pubkey_known is set), so no boot
restructuring was needed despite the initial concern that deferring
Artemis discovery to the idle loop would require one.

Verified: clean build + QEMU boot to [zuse@Hera] ok> on all three
architectures, zero regression to the existing virtio-blk/Zuse-thumbdrive
attach path.

Steps 3 (genesis-stamping onto disk/artemis.img) and 4 (growable
production log-persistence region) not yet started.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01EXieurDfDSsDFdnSyusuWo
This commit is contained in:
Robert Allan James
2026-09-13 07:16:48 -04:00
co-authored by Claude Sonnet 5
parent a8b16d41da
commit 29b6789860
8 changed files with 27473 additions and 8 deletions
+192
View File
@@ -0,0 +1,192 @@
/*
StarForth — Steady-State Virtual Machine Runtime
Copyright (c) 20232025 Robert A. James
All rights reserved.
This file is part of the StarForth project.
Licensed under the StarForth License, Version 1.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at:
https://github.com/star.4th@proton.me/StarForth/LICENSE.txt
This software is provided "AS IS", WITHOUT WARRANTY OF ANY KIND,
express or implied, including but not limited to the warranties of
merchantability, fitness for a particular purpose, and noninfringement.
See the License for the specific language governing permissions and
limitations under the License.
*/
/**
* artemis_sig.h - Artemis disk signature format (FABRIC-3.md §XXVI follow-on,
* 2026-09-13)
*
* Identifies Artemis's own disk, distinct from an identity thumbdrive's
* homeblocks_sig_t -- needed once Artemis's disk stops being found by a
* hardcoded PCI virtio-blk vendor/device ID scan (QEMU-only; real hardware
* has no reason to expose a virtio-blk PCI device at all, since virtio is a
* paravirtualization standard, not something a physical storage controller
* speaks) and starts being discovered generically instead, the same way
* WIREBIND already discovers identity thumbdrives -- by content signature,
* not by which bus happened to present the device. Without a distinct
* signature, generic discovery on real hardware (where an identity
* thumbdrive and Artemis's own disk could both be attached as USB-MSC
* devices simultaneously) would have no way to tell them apart.
*
* Mirrors homeblocks_sig_t's own structural convention exactly (magic +
* version + CRC, one 4KiB devblock, same devblock-1 fixed location) --
* a sibling format, not a field bolted onto homeblocks_sig_t itself:
* homeblocks_sig_t's own header comment already states it's "deliberately
* narrow in scope" (identity-drive fields only, no spare room -- its
* padding is computed to fill exactly 4096 bytes), and Artemis's disk is
* conceptually a different kind of thing (one dedicated fleet-owned device,
* not one of many candidate identity drives), not a variant of the same one.
*
* Reserves offset/size pointers to the growable per-VM log-persistence
* region (FABRIC-3.md §XXVI follow-on's own log-record work), the same way
* homeblocks_sig_t reserves pointers to where the cert and identity source
* attach -- this format doesn't need revisiting when that region's own
* internal layout is designed.
*/
#ifndef STARKERNEL_ARTEMIS_SIG_H
#define STARKERNEL_ARTEMIS_SIG_H
#include <stdint.h>
#ifdef __cplusplus
extern "C" {
#endif
/*===========================================================================
* Magic Field Packing -- same bit layout convention as HOMEBLOCKS_SIG_PACK
*
* bits 0..31 : 'ARTM' (0x4D545241 little-endian) -- distinct from
* homeblocks_sig_t's 'LAHB', so a generic scan can tell an
* Artemis disk apart from an identity thumbdrive by content
* alone, regardless of which bus either was found on.
* bits 32..39 : version (0 for v0)
* bits 40..63 : reserved (zero)
*===========================================================================*/
#define ARTEMIS_SIG_MAGIC 0x4D545241ULL /* 'ARTM' */
#define ARTEMIS_SIG_VERSION_0 0
#define ARTEMIS_SIG_PACK(ver) \
(ARTEMIS_SIG_MAGIC | ((uint64_t)(ver) << 32))
#define ARTEMIS_SIG_GET_MAGIC(m) ((uint32_t)((m) & 0xFFFFFFFFULL))
#define ARTEMIS_SIG_GET_VERSION(m) ((uint8_t)(((m) >> 32) & 0xFF))
/* Same devblock-1 (forth-block 4) convention as HOMEBLOCKS_SIG_START_FBLOCK
* -- devblock 0 stays reserved for the block-subsystem's own generic
* 'STFR'/v2 volume header (block_subsystem.h), same reasoning as
* homeblocks_sig.h's own comment on this. No collision risk with an
* identity thumbdrive's own homeblocks_sig_t at the same devblock offset --
* they are different physical/virtual devices entirely. */
#define ARTEMIS_SIG_START_FBLOCK 4u
/*===========================================================================
* artemis_sig_t - Artemis disk signature header (exactly one 4KiB devblock)
*===========================================================================*/
typedef struct {
uint64_t magic; /* ARTEMIS_SIG_PACK(...) */
uint8_t disk_uuid[16]; /* Mirrors homeblocks_sig_t's drive_uuid --
* one Artemis disk exists today, but costs
* nothing to future-proof the same way. */
uint64_t genesis_time_ns; /* Monotonic timestamp when this signature
* was first stamped (the one-time genesis
* step, not every boot). */
uint64_t metadata_devblocks; /* Size of the metadata region at the start
* of this raw device (sig header + log
* region), in 4KiB devblocks -- everything
* past this is Artemis's own general
* block-storage pool, same "no partition
* boundary" convention homeblocks_sig_t
* uses for an identity's own pool. */
uint32_t log_region_offset; /* Devblock offset where the growable
* per-VM log-persistence region starts;
* 0 = not yet allocated. */
uint32_t log_region_devblocks; /* Current reserved size of the log
* region, in devblocks -- grows over
* time (same growable-reservation
* mechanism the metadata fence design
* already uses elsewhere), not a single
* fixed guess made once at genesis. */
uint64_t hdr_crc; /* Computed over every field above this
* one, same boundary/discipline as
* homeblocks_sig_compute_crc(). */
/* Padding to keep the header exactly one 4KiB devblock. */
uint8_t _pad[4096 - (
8 + /* magic */
16 + /* disk_uuid */
8 + /* genesis_time_ns */
8 + /* metadata_devblocks */
4 + 4 + /* log_region_offset, log_region_devblocks */
8 /* hdr_crc */
)];
} artemis_sig_t;
/* C99-portable compile-time size assertion (no _Static_assert -- that's
* C11), same discipline homeblocks_sig.h's own check uses. */
typedef char artemis_sig_size_check[(sizeof(artemis_sig_t) == 4096) ? 1 : -1];
/*===========================================================================
* Signature check (mirrors homeblocks_sig_result_t exactly)
*===========================================================================*/
typedef enum {
ARTEMIS_SIG_OK = 0, /* magic, version, and crc all check out */
ARTEMIS_SIG_BLANK, /* magic does not match -- blank, foreign, or
* an identity thumbdrive (different magic) */
ARTEMIS_SIG_BAD_VERSION, /* magic matches, version unrecognized */
ARTEMIS_SIG_BAD_CRC, /* magic+version match, crc fails -- corrupt
* or tampered */
ARTEMIS_SIG_READ_ERROR /* could not read from the device at all */
} artemis_sig_result_t;
/* Forward-declared, not included here -- same reasoning as
* homeblocks_sig.h's own forward declaration of struct blkio_dev. */
struct blkio_dev;
/*
* artemis_sig_check - Read and verify the Artemis disk signature header.
* Mirrors homeblocks_sig_check()'s own contract exactly (same forth-block
* read pattern, same "starting block is a caller-supplied parameter"
* separation of concerns).
*
* @param dev Open block device to read from.
* @param sig_start_fblock First of 4 consecutive forth-blocks holding the
* 4KB header -- ARTEMIS_SIG_START_FBLOCK for every
* real caller today.
* @param out_sig On ARTEMIS_SIG_OK, populated with the verified
* header. Left unspecified on any other result.
* @return ARTEMIS_SIG_OK, or the specific reason for refusal.
*/
artemis_sig_result_t artemis_sig_check(struct blkio_dev *dev,
uint32_t sig_start_fblock,
artemis_sig_t *out_sig);
/*
* artemis_sig_compute_crc - CRC-64 over every field of `sig` up to but not
* including hdr_crc itself and the trailing padding. Exposed publicly for
* the same reason homeblocks_sig_compute_crc() is: both the check and the
* future genesis-stamping step need the identical computation.
*
* @param sig Header to checksum. hdr_crc and _pad are not read.
* @return The CRC-64 value that hdr_crc should hold for `sig` to verify.
*/
uint64_t artemis_sig_compute_crc(const artemis_sig_t *sig);
#ifdef __cplusplus
}
#endif
#endif /* STARKERNEL_ARTEMIS_SIG_H */