Commit Graph

  • 4037e746aa docs: storage remount-on-replug is bench-pending, not bench-verified main Daniel Samson 2026-08-10 19:35:04 +01:00
  • 4dfb5012c0 docs: the removal lifecycle closes — three triggers, one path (S5) Daniel Samson 2026-08-10 06:08:56 +01:00
  • 061eb7c004 volume-manager: drop the medium-event dedup — it only ever misfired (S5 review) Daniel Samson 2026-08-10 05:56:06 +01:00
  • 5dc966838a volume-manager: rebuild a volume when its storage driver dies (S5) Daniel Samson 2026-08-10 05:21:48 +01:00
  • 700452dc4e volume-manager: consume medium_changed — the second removal trigger (S5) Daniel Samson 2026-08-10 04:58:54 +01:00
  • 0faa0fd21b service: on_buffered_message for pushed events (S5) Daniel Samson 2026-08-10 04:43:27 +01:00
  • c47215821c docs: the second engine is built — exFAT + proven harness reuse (S4) Daniel Samson 2026-08-10 04:15:03 +01:00
  • 6ddb08091d exfat: adversarial-review fixes — overflow safety, sparse gaps, dir size, big-image bitmap (S4 step 9) Daniel Samson 2026-08-10 04:14:54 +01:00
  • 77b64229c2 exfat: the in-VM mount + mutation drill (S4 step 8) Daniel Samson 2026-08-10 03:44:29 +01:00
  • 90906bcefe exfat: cross-engine discrimination + the exfat-test fixture (S4 step 7) Daniel Samson 2026-08-10 03:32:55 +01:00
  • 2c2745e9e5 volume-manager: recognize exFAT and route it by content (S4 step 6) Daniel Samson 2026-08-10 03:28:08 +01:00
  • 2a6d604577 tools: make-exfat-image.py — a real exFAT image builder (S4 step 5) Daniel Samson 2026-08-10 03:22:19 +01:00
  • e81a4e6f1d exfat: the service + build wiring (S4 step 4) Daniel Samson 2026-08-10 03:18:00 +01:00
  • e240341bfb exfat: the engine write path (S4 step 3) Daniel Samson 2026-08-10 03:11:05 +01:00
  • 62eb2a748a exfat: the engine read path (S4 step 2) Daniel Samson 2026-08-10 03:02:58 +01:00
  • 56bd2e7678 exfat: the on-disk layout (S4 step 1) Daniel Samson 2026-08-10 02:46:23 +01:00
  • 0b25cd2c94 docs: multi-volume is built — storage architecture + rationale (S3) Daniel Samson 2026-08-10 02:35:43 +01:00
  • d59279422e test: partitioned-image tool + shared-channel multi-volume proof (S3) Daniel Samson 2026-08-10 02:35:32 +01:00
  • bf9f8560c6 fat/harness: filesystems coexist without the shared vfs name; two-volume proof (S3) Daniel Samson 2026-08-10 02:10:06 +01:00
  • da7dcce64e kernel/vfs: raise the mount ceiling for N volumes; refuse (not drop) a full table (S3) Daniel Samson 2026-08-10 01:52:47 +01:00
  • 9750db14da fat: install the FHS boot rewrites only on the system volume (S3) Daniel Samson 2026-08-10 01:37:32 +01:00
  • b2a5a0a3c6 volume-manager: adopt every device, a filesystem per partition (S3) Daniel Samson 2026-08-10 01:26:47 +01:00
  • 7efe7b72d8 volume-manager: N-volume device+volume tables, behavior-preserving (S3) Daniel Samson 2026-08-10 01:13:52 +01:00
  • d4b544d66b volume-manager: partition.allVolumes — every partition, not just the first (S3) Daniel Samson 2026-08-10 00:57:56 +01:00
  • 6d4992ae02 docs: storage — the mount map is built; the path is the id (S2) Daniel Samson 2026-08-10 00:49:32 +01:00
  • df61693065 fat: mount at the id-path from argv; migrate /volumes/usb -> id-path (S2) Daniel Samson 2026-08-10 00:47:19 +01:00
  • f1e79d0eeb volume-manager: the volumes query verb — read a volume's id, path, and label (S2) Daniel Samson 2026-08-10 00:17:22 +01:00
  • 167e9c7a9e volume-manager: load the mount map; pick binary by signature, compose the id-path (S2) Daniel Samson 2026-08-10 00:12:33 +01:00
  • 5bfdb75e12 volume-manager: the id-path deriver + volumes.csv override (S2) Daniel Samson 2026-08-09 23:55:08 +01:00
  • 3fb8a9b96f volume-manager: content signature + the filesystems.csv map (S2) Daniel Samson 2026-08-09 23:47:22 +01:00
  • ea8ccf65d0 volume-manager: cover the GPT non-128 entry-size offset path (S1 review) Daniel Samson 2026-08-09 23:37:52 +01:00
  • aca3d5855a volume-manager: S1 close-out — partition fixtures in the root aggregate; docs (S1) Daniel Samson 2026-08-09 23:26:12 +01:00
  • 5637d0e5fc device-manager: the entries-per-reply test asserts 7 (the current shape), not 10 Daniel Samson 2026-08-09 23:26:12 +01:00
  • 48ab12e262 volume-manager: the FAT volume serial + label is the identity (rung 3) (S1) Daniel Samson 2026-08-09 23:22:04 +01:00
  • 020e31bc8f volume-manager: GPT parsing — the partition GUID is the id, the name is the label (S1) Daniel Samson 2026-08-09 23:15:23 +01:00
  • c81120ef0f volume-manager: partition parser takes a SectorReader; identity is a tagged Identity (S1) Daniel Samson 2026-08-09 23:06:23 +01:00
  • 4f9196c03e docs: storage plan — fix two id-path leftovers (DANOS/label->hex) Daniel Samson 2026-08-09 22:55:10 +01:00
  • eff95416d0 docs: storage plan — path is the id, label is queryable display metadata Daniel Samson 2026-08-09 22:54:36 +01:00
  • addd264880 docs: storage plan — volumes named by identity; exFAT implemented in full Daniel Samson 2026-08-09 22:42:56 +01:00
  • fca41b351e docs: the storage-stack completion plan (S1-S5) Daniel Samson 2026-08-09 22:34:00 +01:00
  • bf0595763e docs: flip stale status markers across the tracks (audit found 23) Daniel Samson 2026-08-09 22:04:17 +01:00
  • 8216be991d docs: correct the storage docs' V0-V4 status — the V5 flip missed several Daniel Samson 2026-08-09 21:19:29 +01:00
  • 68e65803eb volume-manager: the removal comment says lazy retirement, not an eager sweep Daniel Samson 2026-08-09 20:56:51 +01:00
  • af47d41989 volume-manager: removal supersedes a pending restart in the poll Daniel Samson 2026-08-09 20:27:58 +01:00
  • 5e89b111cf docs: flip the storage-architecture status markers the V0-V4 track made real Daniel Samson 2026-08-09 20:20:20 +01:00
  • e3ec9fa668 volume-manager: try every mass-storage entry, watch the one we opened Daniel Samson 2026-08-09 20:08:30 +01:00
  • 9e67a74232 volume-manager: the removal lifecycle — a pulled stick unmounts (V4) Daniel Samson 2026-08-09 19:41:04 +01:00
  • b9058fe020 file-system: the mount-failure log has no scratch buffer (orphaned bounds fix) Daniel Samson 2026-08-09 19:40:39 +01:00
  • 7c6ed2ca09 volume-manager: harden the probe and supervision from the V3 review Daniel Samson 2026-08-09 18:50:01 +01:00
  • a67a7015bf docs: record the V3c resequencing — removal lifecycle first, multi-volume follows Daniel Samson 2026-08-09 18:30:16 +01:00
  • 301bdcaf5b volume-manager: the flip — fat is spawned, confined, and handed its channel (V3b) Daniel Samson 2026-08-09 18:28:44 +01:00
  • d56b1b81c0 volume-manager: discovery and probe — V3a Daniel Samson 2026-08-09 18:10:43 +01:00
  • bc67771bfd block: reclaim range slots on death, and give confinement one controller Daniel Samson 2026-08-09 17:58:47 +01:00
  • 89d4592777 block: close the range-clamp overflow — a confined caller could wrap into the neighbour Daniel Samson 2026-08-09 17:40:45 +01:00
  • 7af65697cc block: medium presence — the medium_changed event and usb-storage as publisher (V2b) Daniel Samson 2026-08-09 17:36:57 +01:00
  • 73fbbd3922 docs: record the V2 sequencing — medium_changed emitted in V2b, tested in V4 Daniel Samson 2026-08-09 17:27:43 +01:00
  • c37402891a test: block-range — the discrimination fixture for range confinement (V2a) Daniel Samson 2026-08-09 17:25:43 +01:00
  • f1bdce25e0 block: per-sender range confinement — V2a mechanism Daniel Samson 2026-08-09 17:13:26 +01:00
  • 7ea54a84e5 file-system: mount-failure diagnostics go to the kernel ring, not std.log Daniel Samson 2026-08-09 17:06:13 +01:00
  • d63a008148 file-system: extract the serving harness from fat — V1 Daniel Samson 2026-08-09 16:55:48 +01:00
  • b59f981c58 docs: decision 4 settled — a loop with an open decision is not a loop Daniel Samson 2026-08-09 16:34:04 +01:00
  • 60b41c0e82 kernel: mounts have owners — V0 of the volume-manager plan Daniel Samson 2026-08-09 16:16:14 +01:00
  • 451abba000 docs: the volume-manager plan — V0 unmount ownership through V5 close-out Daniel Samson 2026-08-09 16:09:20 +01:00
  • 9da63e81e2 docs: the boot volume — identity is what makes yanking it survivable Daniel Samson 2026-08-09 16:01:03 +01:00
  • ada251150a docs: volume identity, and volumes.csv as danos's fstab Daniel Samson 2026-08-09 15:59:38 +01:00
  • 728b436d0f docs: the storage rationale lives with the architecture it justifies Daniel Samson 2026-08-09 15:49:12 +01:00
  • 092817ba2e docs: transport generality and the media-presence event Daniel Samson 2026-08-09 15:36:59 +01:00
  • a44b397bed protocols: attach gets its reverse — block detach, usb-transfer dma_detach Daniel Samson 2026-08-09 15:18:21 +01:00
  • fa54ef6915 docs: the volume lifecycle is enforced, not described Daniel Samson 2026-08-09 15:13:34 +01:00
  • fae616fa5a docs: the storage architecture — layers, boundaries, and who does what when media leaves Daniel Samson 2026-08-09 15:08:45 +01:00
  • 5a8a2a4d7e docs: the storage-stack discussion — block, volumes, filesystems, against the survey Daniel Samson 2026-08-09 14:47:35 +01:00
  • d4f8dc51b9 docs: the hot-plug matrix is complete — 124/124 Daniel Samson 2026-08-09 14:19:39 +01:00
  • 36ca3f98b3 test: H4 + H5 — a moved device is a new device, and generations do not leak Daniel Samson 2026-08-09 14:11:05 +01:00
  • c3604429d4 test: H3 — a nested hub tree yanked whole rebuilds whole (hot-plug matrix) Daniel Samson 2026-08-09 14:09:19 +01:00
  • 77e7001878 usb: a hub yanked from a root port takes its subtree with it (H2) Daniel Samson 2026-08-09 14:08:15 +01:00
  • a6a3402d92 test: H1 — root-port unplug and replug (hot-plug matrix) Daniel Samson 2026-08-09 14:03:02 +01:00
  • d565a6b845 docs: the hot-plug matrix plan — unplug anything, replug anywhere Daniel Samson 2026-08-09 13:58:29 +01:00
  • 4a2df587bb establishment: unplug reaps like death, so a replug rebinds Daniel Samson 2026-08-09 13:51:07 +01:00
  • 3c9f454398 establishment: the two-controller proof, and the docs catch up Daniel Samson 2026-08-09 12:48:33 +01:00
  • 8710944a92 establishment: a reporter's death reaps its subtree, and the re-report rebuilds it Daniel Samson 2026-08-09 12:34:13 +01:00
  • 1d7850239d establishment: block stops being a name, and enumerate learns to page Daniel Samson 2026-08-09 12:24:25 +01:00
  • d603d40b5c establishment: usb-transfer stops being a name Daniel Samson 2026-08-09 11:48:31 +01:00
  • 77fe4d220e establishment: the mechanics — reply capabilities, helloExchange, lineage routing Daniel Samson 2026-08-09 11:39:29 +01:00
  • 0d5a7394ef docs: establishment planes — the design and the two-seam plan Daniel Samson 2026-08-09 11:26:53 +01:00
  • 5930c9653c docs: device-authority as built — spawn carries the device, the loan, confinement rebuilt on return Daniel Samson 2026-08-09 10:17:46 +01:00
  • 981ff7cd25 Merge the audit fixes: unlocked give paths, restart re-confinement, AMD-Vi store bit Daniel Samson 2026-08-09 10:03:42 +01:00
  • 5cca580066 kernel: AMD-Vi asked for an interrupt where it meant a store Daniel Samson 2026-08-09 09:57:55 +01:00
  • 35f43057f4 kernel: the give paths take the lock, and a give confines afresh after a death Daniel Samson 2026-08-09 09:57:55 +01:00
  • 72807c20e4 Merge the bounds track: the numbers we invented are gone Daniel Samson 2026-08-08 23:24:25 +01:00
  • 0eb2420690 device-manager: delete the delegated-set scaffolding Daniel Samson 2026-08-08 23:01:29 +01:00
  • df9c1ed827 device-manager: hold the seeded hardware so none is left lying around Daniel Samson 2026-08-08 22:42:42 +01:00
  • ba195fa0a2 kernel: claim refuses delegated hardware — and E2 had already closed the hole Daniel Samson 2026-08-08 22:31:04 +01:00
  • 1a1d92cba9 kernel: a grant is a loan — a dead borrower returns the device to its lender Daniel Samson 2026-08-08 22:21:25 +01:00
  • 4ca57fc37e kernel: record who gave each device away Daniel Samson 2026-08-08 22:12:01 +01:00
  • ca1126537d docs: Run 3 — close the claiming hole, six steps, no open questions Daniel Samson 2026-08-08 22:01:49 +01:00
  • 337b981c07 docs: reorder the Run 2 table to the plan order and note the executed sequence Daniel Samson 2026-08-08 21:58:53 +01:00
  • b7d97ebb5d acpi: discovery is handed its node like every other driver Daniel Samson 2026-08-08 21:54:45 +01:00
  • 6b3a381626 ps2: one instance, every node the machine has Daniel Samson 2026-08-08 21:33:35 +01:00
  • 7d8aa51234 kernel: maximum_children_per_parent is gone Daniel Samson 2026-08-08 21:12:20 +01:00
  • 3ae541214f iommu: assert directly that a confinement moves with its device Daniel Samson 2026-08-08 19:57:36 +01:00