libfreemkv/CHANGELOG.md

Changelog

0.26.1 (2026-05-22)

Added

• AACS resolver path 5 — KEYDB unit-keys direct fallback. When a KEYDB entry for the disc has no VUK field but does have pre-decrypted unit keys, the resolver now consumes those unit keys directly instead of treating the entry as unusable. Covers ~4,572 entries in the public keydb (≈2.5% of the database, heavily skewed toward MKBv76+ UHD discs that DVDFab/FindVUK can no longer extract VUKs for). Surfaces as KeySource::KeyDbUnitKeys; AacsState::vuk and ResolvedKeys::vuk are None on this path because no VUK exists to return. Partial CPS-unit coverage is rejected so a disc is never half-decrypted.

Changed

• AACS resolver path order reordered root-to-leaf. Paths now run in derivation-strength order: device-key (1) → processing-key (2) → KEYDB-derived from MK+VID (3) → KEYDB VUK direct (4) → KEYDB unit keys direct (5). Previous order was leaf-first (KEYDB hash lookup, then derivation). Resolution runs once per disc, so the few extra milliseconds of MKB walking when a KEYDB VUK would have answered are invisible against the rip itself; the new order matches the AACS derivation hierarchy. KeySource enum variants are reordered to match.
• AacsState::vuk type is now Option<[u8; 16]> (was [u8; 16]). Required by path 5, which has no VUK. Public-API break.
• ResolvedKeys::vuk type is now Option<[u8; 16]> (was [u8; 16]). Same rationale.

0.25.14 (2026-05-21)

Changed

• Drive::is_libredrive_active() renamed to Drive::is_raw_read_active(). Same semantics; old name removed. Mirrored on the internal PlatformDriver::is_libredrive_active() trait method (now is_raw_read_active()).
• Error::AacsLibredriveUnsupported renamed to Error::AacsRawReadUnsupported; the underlying numeric code (E7016) is unchanged. The E_AACS_LIBREDRIVE_UNSUPPORTED constant is renamed to E_AACS_RAW_READ_UNSUPPORTED.

No behavioural change — purely a rename pass.

0.25.13 (2026-05-21)

Added

• DrmScheme top-level dispatcher. New drm module with a DrmScheme enum (Css, Aacs10, Aacs20, Aacs21) and a detect + load pair that uniformly handles all four content protection schemes. Replaces the inlined dispatch in disc::encrypt::resolve_encryption and the scattered CSS routing in disc::mod. Both CSS call sites now route through the same entry point.
• AACS 2.1 Media Key Variant framework. New aacs::variants module implementing the Media Key Variant derivation chain (Kp + C → Kmp → ⊕KCD → Kpnew → Km → VUK), Variant-scheme MKB record parsing (record types 0x82 / 0x83), bit-0x02 SoftKCD and bit-0x04 online-challenge detection with dedicated error variants. Wired into DrmScheme::Aacs21 but the dispatcher arm is commented out pending validation against a Variant-scheme disc. Per-manufacturer Key Correction Data must be supplied by the integrator; KEY_CORRECTION_DATA_PLACEHOLDER is the empty placeholder slot.
• AacsVersion enum. Replaces the aacs2: bool field on ContentCertificate, UnitKeyFile, and ResolvedKeys. parse_unit_key_ro and parse_content_cert now take/emit the enum. resolve_keys is split into resolve_keys_v1, resolve_keys_v2, and resolve_keys_v21 (the last not reachable from the dispatcher today).

Fixed

• Libredrive raw-read VID shortcut deleted. v0.25.11 introduced a do_handshake branch that, on libredrive-active drives, skipped the AACS cert handshake and issued READ_DISC_STRUCTURE format 0x80 with AGID=0 directly. The hypothesis was that unlocked drives would serve VID without auth. Empirical test (BU40N
  • UHD disc, 2026-05-21) showed the drive returns 0x05 / 0x6F / 0x02 (ILLEGAL_REQUEST / Copy protection key exchange failure: KEY NOT ESTABLISHED) to that CDB regardless of drive-unlock state. The AACS spec requires a successful REPORT_KEY / SEND_KEY exchange to establish an AGID before format 0x80 returns VID; that requirement is enforced by the drive itself and isn't bypassed by libredrive firmware. The shortcut fired for every libredrive-active drive, so v0.25.11 / v0.25.12 MOVIE scans were stuck at E7017 instead of progressing to the real wall (no DK walks MKB v77).
• Disc::do_handshake now always routes through do_handshake_cert. Drive::is_libredrive_active() and the Mt1959 MMkv+LbDr marker detection are kept as informational signals (logged in the handshake_entry warn line) but no longer steer the auth path.
• read_volume_id_libredrive deleted (~50 LOC).

The corollary: AACS resolution on HRL-burned drives + UHD discs now fails honestly. Either cert auth succeeds (drive unlock may or may not bypass the HRL — that's the new empirical question) and we hit the actual DK wall (E7018 "No DK that walks this MKB" for v77+ UHD without a v77+ DK in keydb), or cert auth fails and we surface E7015. Both are real verdicts; E7017's previous spurious dispatch is gone.

0.25.12 (2026-05-21)

No libfreemkv source changes — unified sync bump for autorip's aacs_failure_message two-line wording rewrite. See the autorip v0.25.12 release for details.

0.25.11 (2026-05-21)

Added

• Libredrive raw-read VID path. When the Mt1959 unlock response confirms both the active-mode (MMkv) and mode-ID (LbDr) markers, Drive::is_libredrive_active() returns true and do_handshake skips the AACS cert dance entirely — VID is retrieved via READ_DISC_STRUCTURE format 0x80 with AGID=0, and bus encryption is already off. This is what unblocks UHD ripping on drives whose leaked host cert is on the AACS HRL.
• New Error variants for finer-grained AACS failure reporting: AacsHostCertRejected (E7015), AacsLibredriveUnsupported (E7016), AacsVidUnavailable (E7017), AacsMkUnavailable (E7018), AacsVukNotInKeydb (E7019). Lets CLIs/UIs render which piece of the AACS chain failed instead of always saying "no keys."

Fixed

• validate_processing_key now matches libaacs _validate_pk exactly: XORs uv into mk[12..16] (was omitted), AES-decrypts mk_dv and checks the 01 23 45 67 89 AB CD EF magic (was AES-encrypt + 12-zero check). Pre-fix, every non-zero-uv processing key was rejected — i.e. essentially every real disc.
• mkb_find_cvalues now prefers record type 0x07 (AACS 2.x) and falls back to 0x05 (AACS 1.0), so the walker handles both generations without an out-of-band version flag.
• resolve_keys short-circuits paths 2/3/4 when VID is the zero sentinel — saves cycles and emits an honest "VID unavailable" log instead of the misleading "all paths failed."
• AES-CMAC VID verification gains NIST SP 800-38B KAT + round-trip
  • mutation + all-zero-rejection tests.

Removed

• Built-in AACS keys (added in 0.25.9). src/aacs/builtin_keys.rs deleted; KeyDb::with_builtins, KeyDb::load_or_builtins, KeyDb::merge_from, and internal dedup helpers gone. The compiled-in shortcut was a slim convenience that didn't move the hard problem (no v77+ DKs) and added a maintenance surface.
• Operator plugin slot at ~/.config/freemkv/local_keys.cfg (added in 0.25.9). local_plugin_path and KeyDb::merge_local_plugin removed. Single source: the main keydb.cfg.
• KeyDb::load reverts to the pre-0.25.9 form — read the file or return I/O error; no layering, no fallback.
• Disc::scan reverts to surfacing KeydbLoad { path: "<no keydb in search paths>" } for an encrypted disc with no keydb — same sentinel autorip's message switch already handles.

CSS player keys in src/css/auth.rs remain compiled in; they're 1999-era public inputs separate from the AACS pipeline and were never part of the 0.25.9 additions.

0.25.9 (2026-05-20)

Fixed

• MKB record-type identification. mkb_find_mk_dv and mkb_version had their record-type constants swapped. The Verify Media Key record is Type 0x81 (AACS 1.0) or Type 0x86 (AACS 2.0/2.1), not 0x10. The Type-and-Version record is Type 0x10, not 0x81. PK and DK derivation paths therefore failed silently on every disc, masking how often the fallback paths could have succeeded. Fixed; mk_dv extracted at offset 4 of the verify record, MKB version read at offset 8 of the Type 0x10 record body. Tests added for both forms.

Added

• Built-in AACS keys. Four device keys (covering MKB v01-v82+) and three processing keys (covering v63-v68) are now compiled into libfreemkv directly. DVDs and Blu-rays (AACS 1.0) decrypt with zero external files. Combined with the existing 31 CSS player keys in css/auth.rs, the library is self-sufficient for all DVD and AACS 1.0 content.
• Operator plugin slot. ~/.config/freemkv/local_keys.cfg is loaded automatically (same format as keydb.cfg) and layered additively on top of the built-ins and the main keydb.cfg. For operators who derive their own keys, this is the additive surface to drop them into — main keydb.cfg from upstream auto-update stays separate and overwritable.
• KeyDb::with_builtins(), KeyDb::load_or_builtins(), KeyDb::empty() constructors.

Changed

• Disc::scan no longer errors when keydb.cfg is absent. With built-ins covering DVD/BD, missing keydb is treated as "no UHD keys available" — the AACS 2.0 / UHD code path surfaces a specific error only when the disc actually needs keys that aren't in built-ins, main keydb, or local plugin.

0.25.8 (2026-05-20)

Changed

• Synchronized to autorip 0.25.8. No libfreemkv code changes — unified versioning bump only. autorip 0.25.7 image was DOA (missing +x on the binary); see autorip changelog.

0.25.7 (2026-05-20)

Fixed — BU40N firmware wedge on KEYDB-miss discs

disc::encrypt::do_handshake used to fire up to 16 AACS authenticate attempts back-to-back with no pause between them. Each attempt is 5-10 SCSI REPORT_KEY/SEND_KEY exchanges. On a disc whose host cert isn't in our KEYDB (or one the drive rejects), that's 80-160 SCSI commands hammered at the drive in a few hundred milliseconds — and the BU40N (plus most consumer optical drives) responds by entering a fast-fail firmware wedge state where every subsequent CDB returns ILLEGAL_REQUEST/INVALID_FIELD_IN_CDB (sense 0x05/0x24) until the drive is physically power-cycled.

Live wedge event 2026-05-20 during a UHD disc scan (KEYDB miss) confirmed the diagnosis and motivated this fix.

Defence-in-depth:

• MAX_CERT_ATTEMPTS capped at 3 (was 16). If three different host certs all fail, more won't help — the drive doesn't have a match.
• 1-second sleep between attempts. Gives the drive's firmware time to recover internal state between auth challenges.
• Bail immediately on any sense code with sense_key = 0x05 (ILLEGAL_REQUEST). The drive isn't merely rejecting our cert — it's saying "I won't talk to you anymore" — so trying more certs would deepen the wedge.

0.25.6 (2026-05-20)

Changed

• Synchronized to autorip 0.25.6 release. No behavioural changes.

0.25.5 (2026-05-20)

Changed

• Synchronized to autorip 0.25.5 release. No behavioural changes.

0.25.4 (2026-05-20)

Changed

• Synchronized to autorip 0.25.4 release. No behavioural changes in the library — the v0.25.4 release ships the autorip-side in-container NFS mount option.

0.25.3 (2026-05-19)

Changed

• Synchronized to autorip 0.25.3 release. No behavioural changes in the library — the v0.25.3 release ships the autorip-side parallel mux pipeline.

0.25.2 (2026-05-19)

Fixed

• DTS-HD codec ID — MkvTrack::audio now emits A_DTS/MA for DTS-HD MA and A_DTS/HR for DTS-HD HR instead of mislabelling both as plain A_DTS. Strict players (Plex transcoder, some hardware decoders, AV receivers) reject lossless DTS-HD MA payload when the track advertises A_DTS because the ID implies the 1.5 Mbps core-only bitstream.
• PGS subtitle BlockDuration — the PGS parser is now stateful: it pairs each display PCS with the following empty/clear PCS to compute a duration, and the MKV muxer emits a BlockGroup + BlockDuration for subtitles that carry one. Without this the last bitmap lingered on screen until the next display set replaced it (or until end of file).

Changed

• codec::Frame gains duration_ns: Option<u64> (set by parsers that can compute one; currently only PGS).
• pes::PesFrame gains duration_ns: Option<u64> (in-memory only; not part of the on-wire serialization).
• MkvMuxer::write_frame now takes a final duration_ns: Option<u64> parameter. When Some, the frame is emitted as a BlockGroup with BlockDuration instead of a SimpleBlock.

0.25.1 (2026-05-19)

New — autorip event_fn plumbing

• PrefetchedSectorSource::new_with_events(reader, extents, batch, halt, event_fn) — same producer-thread pipeline as new(), plus an optional callback fired from the producer with BytesRead events after every successful batch. Lets the autorip multipass
  • resume mux paths drive their progress UI from the highway without polling the consumer side.
• build_iso_pipeline gains an event_fn parameter so callers can wire the same callback through one ctor.

Changed

• pes::Stream trait: new default errors() -> u64 method (default returns 0). Lets Box<dyn Stream> callers query the skip-on-error counter without downcasting. DiscStream overrides to surface its existing errors field.

Removed (breaking)

• DiscStream::new_pipeline and DiscStream::read_pipeline deleted. All file-backed mux now uses build_iso_pipeline → PipelinedPesStream. DiscStream is now the single-threaded inline reader used by autorip's live-drive single-pass path only.
• DiscStream::demux_thread and demux_rx fields removed.
• M2tsStream::open deleted; Mode::Read variant removed. m2ts:// URLs go through the internal build_m2ts_pipeline helper in mux/resolve.rs → PipelinedPesStream. M2tsStream is a write-only sink now.

0.25.0 (2026-05-19)

New — the freemkv mux throughput "highway"

Three-stage pipelined PES read path. Read+decrypt runs on a producer thread, M2TS demux runs on a second thread, codec parse runs on the caller's thread. Communication between stages is via bounded crossbeam_channel with a recycled buffer pool — no allocations or memcpys in the steady-state hot loop.

Throughput on the test bed (a UHD disc, 62 GiB ISO → null://, single-thread caller):

	MB/s
0.23.2 baseline	60
+ memchr SIMD HEVC start-code scan	69
+ ts.feed no-copy boundary	72
+ PrefetchedSectorSource (producer)	124
+ DemuxThread (3-stage pipeline)	135
+ zero-copy recycled buffer pool	148
+ 16 KiB initial PesAssembler buffer	162
+ mimalloc allocator in freemkv CLI	200+
+ warm cache	660

The new public API:

• libfreemkv::PrefetchedSectorSource — wraps any SectorSource, spawns a producer thread, exposes recycled-buffer channels.
• libfreemkv::io::byte_prefetcher::BytePrefetcher — std::io::Read analogue for byte-stream sources (m2ts files, sockets, stdin).
• libfreemkv::mux::demux_thread::DemuxThread — the M2TS demux worker; spawn_zero_copy takes either prefetcher's channels.
• libfreemkv::PipelinedPesStream — the read-side Stream impl that runs codec parse on the caller thread.
• libfreemkv::build_iso_pipeline(reader, title, keys, batch, format, halt) — the canonical ctor that wires all three stages for an ISO file source.

Changed (breaking)

• IsoSectorReader (the naive duplicate of FileSectorSource) deleted. FileSectorSource is the sole file-backed sector source; it carries the SEQUENTIAL fadvise hint, the periodic DONTNEED page-cache eviction, and (new) the per-read readahead() async-prefetch syscall.
• mux::input("iso://...", &opts) now returns a PipelinedPesStream. Function signature input(&str, &InputOptions) -> io::Result<Box<dyn Stream>> is unchanged; callers that treated the return value as Box<dyn Stream> keep working.

Other

• FileSectorSource exposes per-OS prefetch() hooks (readahead(2) on Linux, fcntl(F_RDADVISE) on macOS, no-op on Windows + other).
• AACS decrypt thread-pool: env var renamed from FREEMKV_DECRYPT_THREADS to FREEMKV_THREADS; default raised to cores.clamp(1, 64).
• PesAssembler initial buffer capacity 256 KiB → 16 KiB (avoids the 64-page first-touch fault tax on every PES boundary).
• HEVC / H.264 find_start_code swapped onto memchr::memmem::find (SIMD).

0.18.4 (2026-05-09)

Build / CI hardening — no library code changes

• All cargo invocations in .github/workflows/*.yml now use --locked. Previously cargo build / cargo test could silently re-resolve Cargo.lock if a lock-pinned version wasn't yet on crates.io — masking dependency races between same-tag releases of sibling crates. The 0.18.3 release hit this: autorip's docker image built libfreemkv v0.18.2 because libfreemkv 0.18.3 hadn't published yet at the time autorip's CI ran, and cargo silently fell back to the previous version. With --locked, that race now hard- fails the build with a clear "package X is not in registry" error and we retrigger after the upstream lands.

• cargo publish --locked in the release workflow guards against publishing a lockfile-mismatched release.

0.18.3 (2026-05-09)

Behaviour change

• Disc::titles[0] is now the canonical main feature on branching discs, not the longest playlist. Previously titles were sorted purely by duration_secs descending, which puts a "play-all" virtual playlist (alternate angles / seamless branching) at index 0 on UHDs that ship one — its inflated duration overshoots the real movie. Concrete example: The Amateur (2025) 4K UHD on a 58.5 GB disc has Title 1 = 00020.mpls 4h13m / 92.4 GB / 253 clips (impossible — the size exceeds the disc capacity, proving it's a virtual composite) and Title 2 = 00800.mpls 2h02m / 57.2 GB / 1 clip (the actual film, matching TMDB).

  New sort priority:

  1. Real titles (size_bytes <= capacity_bytes) before virtual composites.
  2. Among real titles, fewer clips first (1-clip wins).
  3. Tiebreak on longer duration first.

  Migration: consumers calling disc.titles.first() / disc.titles[0] automatically get the corrected title — no code changes needed. CLI users invoking freemkv -t 1 disc:// … now hit the actual main feature on branching discs (this was the user- visible bug). On non-branching discs the order is unchanged.

  The comparator is exposed as Disc::canonical_title_order for callers that need to re-sort a custom title set with the same logic.

  Regression tests: disc::tests::canonical_order_* (three cases: branching-UHD, normal disc, clip-count tiebreak).

0.18.2 (2026-05-09)

Bug fixes

• Nav-file scramble during AACS rip (decrypt::decrypt_sectors). 0.18's DecryptingSectorSource decorator broadened the call surface of decrypt_sectors to every sector flowing through sweep, including UDF navigation files (MPLS playlists, CLPI clip-info). The byte-0 heuristic in aacs::is_unit_encrypted correctly fires on m2ts source-packet copy markers but false-positives on any binary file whose first byte happens to have the top 2 bits set — most notably MPLS files (start with 'M' = 0x4D) and CLPI files (start with 'H' = 0x48). decrypt_unit_full already self-checks the result via TS-sync verification and returns false on a misfire, but the chunk had been mutated by then; decrypt_sectors discarded the return value, leaving scrambled bytes in the ISO. Fix: snapshot the chunk before decryption and restore on verification failure (same pattern decrypt_unit_try_keys already used for multi-key discs). Symptom: freemkv info iso://UHD.iso and iso:// → mkv:// returned E6009 NoStreams on freshly-ripped UHD ISOs; affected only the iso-source path, not the disc-source path or the m2ts video payload itself. Regression test: decrypt::tests::nav_file_unit_survives_decrypt_attempt.

• Sweep progress display can regress to zero (Disc::sweep). The consumer-side bytes_good snapshot lags producer-side bytes_done whenever the consumer is behind on draining the work channel. Until the first snapshot arrived, the placeholder branch reported bytes_done correctly; once a stale snapshot landed, the report switched to snap.bytes_good and could regress below bytes_done. Fix: bytes_good_total = max(snap.bytes_good, bytes_done), so the user-visible counter never moves backward.

0.18.1 (2026-05-09)

I/O stack redesign — primitives over orchestration

0.18 reshapes the read/write surface from "library does the multipass dance" to "library hands the caller flat verbs and a few composable primitives." All in-tree consumers (autorip, the freemkv CLI) drove their own multipass loops in 0.18 rounds 1-3; 0.18.1 lands the cleanup with the deprecated names still alive for one minor-version window.

The 0.18 design notes are kept privately — this entry sticks to what changed at the public surface.

Flat verbs

Disc::sweep is the forward Pass 1. Disc::patch is one retry pass over the mapfile. Neither knows about pass index, retry budget, or accept-loss policy — the caller invokes them in whatever sequence its use case dictates. The old multipass-aware Disc::copy dispatcher is deprecated and slated for deletion in 0.18.2; no in-tree caller still uses it.

Trait splits — direction-typed at compile time

• pes::Stream (combined read+write) is split into FrameSource and FrameSink. Calling read() on a write-only sink is now a compile error, not the runtime E9001 (StreamWriteOnly) it used to be.
• SectorReader is split into SectorSource (read) and SectorSink (write). Drive impls SectorSource only; FileSectorSource / FileSectorSink replace FileSectorReader for ISO-backed I/O.
• A blanket impl bridges legacy SectorReader callers onto the new SectorSource so existing code keeps compiling through the deprecation window.

New primitives

• Halt — one cancellation token (cloneable, Arc<AtomicBool> under the hood) replaces the three near-duplicate halt flags scattered through the workspace. Threaded through every long-running loop.
• Pipeline<I, R> + Sink<I> — generic producer/consumer primitive in crate::io. Replaces the bespoke disc/sweep_pipeline.rs and now also drives Disc::patch and the autorip mux loop. DEFAULT_PIPELINE_DEPTH for streaming reads (4); WRITE_THROUGH_DEPTH for write-through patch semantics (1).
• WritebackFile — was crate::io::Writer. The renamed type makes its job explicit: a File wrapper that runs continuous sync_file_range + posix_fadvise(DONTNEED) to keep the kernel dirty-page cache bounded on long sequential writes.
• DecryptingSectorSource<S> — a single decorator wrapping any SectorSource to yield plaintext sectors. One audit surface for AACS / CSS / passthrough; the previous two-site decrypt (sweep producer + DiscStream demux) is gone.

Throughput

The round-2 producer/consumer split is now applied uniformly to sweep, patch, and mux. Mux on NFS-staged UHD measured ~16 MB/s sustained on the test bed (was ~12 MB/s pre-round-2).

Module reorg

• sector/ and io/ are now module directories.
• disc/sweep.rs, disc/patch.rs, and disc/mapfile.rs split out of the monolithic disc/mod.rs.

Renames (no behavior change)

• crate::io::Writer → crate::io::WritebackFile.
• Apply → Flow (Sink::apply return value).
• DEFAULT_DEPTH → DEFAULT_PIPELINE_DEPTH.
• PatchOpts → PatchOptions.

Deprecated (alive in 0.18.1, deletion target 0.18.2)

Disc::copy, pes::Stream, SectorReader, FileSectorReader, CopyOptions, CopyResult, DiscStream::set_halt. Each compiles with a deprecation warning; all in-tree call sites have migrated.

0.17.13 (2026-05-09)

Use crate::io::Writer uniformly for all binary file output

0.17.10 introduced the bounded-cache writeback wrapper (sync_file_range

• posix_fadvise(DONTNEED) per chunk) and 0.17.11 wired it into Disc::sweep. The other two big-write paths in the crate were still opening raw std::fs::File and would have hit the same dirty-page burst pathology against slow / network-attached staging.

This release threads crate::io::Writer through the remaining sites:

• Disc::patch (disc/mod.rs:1981): the ISO file reopened for Pass-N recovery now wraps in Writer before any seek / write. Recovery writes are sparse, but the wrapper costs nothing when there's no chunk crossing — and on heavily-damaged discs it matters as patch accumulates GB of recovered data.
• MKV mux output (mux/resolve.rs:243): BufWriter::with_capacity now wraps Writer::new(file) instead of a raw File. UHD MKVs routinely exceed 70 GB of sequential writes; pre-0.17.13 those bursts went straight to the kernel writeback queue.
• M2TS mux output (mux/resolve.rs:251): same change for parity with the MKV path — anyone using m2ts:// URLs gets it too.

No new public surface. Writer::sync_all() is called from Disc::patch and sweep_pipeline's consumer at end of pass; mux exits via Drop on the wrapping BufWriter, which propagates flush down to Writer's final note_progress (kernel finishes the in-flight chunk on file close — there's no explicit sync_all for mux today, same behaviour as before).

0.17.12 (2026-05-09)

Mapfile time-batched persistence — unblock NFS staging

Pre-0.17.12 every Mapfile::record() wrote the entire mapfile to disk via tempfile-create + buffered-write + atomic-rename. On local LVM that's effectively free (page cache + microsecond-scale renames). On NFS (autorip's intended staging path for centralised media) each record() became three RPCs through the unraid user share's shfs-fuse layer — measured end-to-end at multiple ms each. With ~170 record() calls per second of sustained sweep, the mapfile path alone burned multiple seconds of wall time per real-world second of work, dragging the rip from ~11 MB/s on local to ~1.5 MB/s on NFS.

Fix: time-batch the persistence inside Mapfile.

• record() always updates in-memory state and stats (so callers' stats() reads stay coherent in the same process).
• The write_to_disk() rename only fires when ≥ FLUSH_INTERVAL (1 s) has elapsed since the last persist.
• New flush() method forces an out-of-band persist; called by sweep_pipeline's consumer at end-of-sweep and by Disc::patch at end-of-patch, after the file's sync_all().
• Drop impl best-effort flushes so an early-return / unwind doesn't silently lose pending state.

Crash-safety changes from "lose at most one block" to "lose at most 1 s of recorded progress" — the ISO file's payload bytes are unaffected; only the mapfile's authority over which sectors are already-good is at risk, and a resume re-reads anything Pass 1 had already covered. Acceptable for a 7× throughput recovery on the target deployment.

The internal round_trip_load test now calls flush() before Mapfile::load to read back what the in-memory state asserts. External patch / copy tests are unaffected: patch and sweep_pipeline flush at completion before returning, so any Mapfile::load at the call-site sees fully persisted state.

0.17.11 (2026-05-09)

Sweep producer/consumer split — overlap drive read with file write

Pre-0.17.11 the sweep loop ran strictly serialised: SCSI read → decrypt → seek + write → mapfile.record → next read. The drive idled for the post-read work; throughput capped at the sum of both costs. On a healthy disc that's ~7-12 ms read + ~5-15 ms write/record per 64 KB batch — limiting sustained throughput to ~10-12 MB/s on the test bed (BU40N + UHD inner zone), well below the drive's ceiling of ~14-16 MB/s.

This release decouples them with a producer / consumer split:

• Producer (caller's thread): owns the SectorReader, the entire read_error state machine (Retry, Bisect, SkipBlock, JumpAhead, AbortPass), set_speed damage-zone transitions, halt check, and decrypt. Hands plaintext bytes to the consumer.
• Consumer (one spawned thread): owns the crate::io::Writer and the Mapfile. Receives WorkItem messages and applies the file write + mapfile record per item.
• Channel: bounded mpsc::sync_channel(4) — natural back-pressure via blocking send when consumer falls behind.

While the consumer is writing batch N to disk and updating the mapfile, the producer is already reading batch N+1 from the drive. Steady-state throughput is now bound by the slower of the two pipelines (the drive, on a healthy disc) instead of their sum.

Side effects of the refactor:

• Bisect path now decrypts. Pre-0.17.11 the bisect inner loop wrote raw cyphertext when decrypt=true and a single sector was recovered via single-sector retry — a quiet correctness bug exercised only by the (rare) batch-fail-then-bisect-succeed path on encrypted discs. The new producer-side decrypt covers both the main success path and the bisect inner success path.
• All read_error::ReadCtx state stays single-threaded on the producer (damage window, jump multiplier, consecutive-good count, etc.). No locking added.
• Mapfile remains single-writer (consumer-only). No locking.
• Halt-flag responsiveness unchanged: producer breaks the loop on signal, sends Finish, consumer drains its ≤4 in-flight items and exits within ~1 batch (~12 ms typical).
• BU40N + Initio bridge wedge concern unchanged: still one SCSI command in flight, error-path timing identical, no new retry logic.

New module: src/disc/sweep_pipeline.rs (WorkItem, ProgressSnapshot, ConsumerInputs, spawn_consumer, consumer_loop, send/recv helpers). Public API surface unchanged — Disc::copy / CopyOptions / CopyResult look identical to callers.

Patch (Pass N) is not affected by this release. Patch is bound by drive recovery time (60 s timeouts on bad sectors), not the read↔write serialisation; a similar split there would yield negligible benefit.

0.17.10 (2026-05-09)

Bounded-cache writeback for big sequential writes

Pass 1 sweep speed on a healthy disc previously dipped from ~15 MB/s to ~1 MB/s every ~30 s on a host with default Linux dirty-page settings. Cause: the kernel's vm.dirty_ratio (~20 % of RAM) lets hundreds of MB of dirty pages accumulate, then bursts a flush at 99 % disk utilisation that blocks app writes for ~1 s. Confirmed empirically on the BU40N test bed: dirty pages grew 112 → 563 MB between bursts; lowering vm.dirty_bytes to 64 MB at the host sysctl level eliminated the dips.

This release ships the equivalent inside libfreemkv so users don't need to tune the host kernel:

• New crate::io::Writer — drop-in File wrapper implementing Write + Seek. Wraps a per-platform WritebackPipeline that on Linux schedules sync_file_range(WRITE) + lagging sync_file_range(WAIT_AFTER) + posix_fadvise(DONTNEED) calls in 32 MB chunks, keeping dirty cache bounded at ~64 MB. macOS and Windows ship a no-op stub — their default cache policies don't exhibit the same pathology for our access pattern.
• Disc::sweep wraps its output File in Writer. No changes to the loop body — Writer forwards seek/write_all to File and drives the pipeline transparently.
• Module is purpose-built for any large sequential output (sweep, patch, future mux) — they can adopt crate::io::Writer with a one-line wrapper and inherit the same behaviour.

0.17.7 (2026-05-08)

Sync release — no functional libfreemkv changes

Version bump to keep the four freemkv crates at unified versioning. autorip shipped UI smoothness + audit fixes today (v0.17.6, v0.17.7); libfreemkv carries no code changes, but is bumped + republished to crates.io so downstream libfreemkv = "0.17" consumers stay aligned on the latest patch version. See autorip CHANGELOG for the user-visible work in this release window.

0.17.5 (2026-05-08)

Pass N recovery — kernel block-device fallback + per-range fixes

Live testing on direct-SATA BU40N + AACS-encrypted UHD disc revealed that the v0.17.3 single-shot SCSI READ path matched 0/22 of the small bad-sector LBAs that dd if=/dev/sr0 recovers on the same drive. This release closes that gap and fixes several adjacent bugs that were silently capping recovery.

• /dev/sr0 pread fallback in Drive::read (Linux): when a SCSI READ via /dev/sg* returns Err, fall back to `posix_fadvise(POSIX_FADV_DONTNEED)

  • pread()against the corresponding block device. The kernelsr_moddriver runs ~5 internal retries per command without the per-attempt error-escalation overhead that userspace SG_IO retries pay, which is the source of dd's recovery advantage. End-to-end byte-verification confirms the fallback path returns real disc data (md5-equivalent to fresh dd from the same drive session). The block fd is opened inDrive::openby resolving/sys/class/scsi_generic/sgN/device/block`; best-effort, with no fallback if open fails.

• Disc::patch per-range watchdog fix: when a range hit MAX_RANGE_SECS, the old code did wedged_exit = true; break 'outer; — a single slow range killed the entire patch. Now break; (skip this range, advance the outer for loop). Pre-fix, patch was dying after 4 sectors of the first slow range and never reaching the other 46.

• Per-sector range budget: replaced flat MAX_RANGE_SECS = 180 with range_budget_secs = (range_sectors × SECONDS_PER_SECTOR).min(RANGE_BUDGET_CAP_SECS). Tiny ranges exit fast (1-sector range = 25 s); medium ranges get proportional time (51-sector range ≈ 1275 s); large ranges still bounded by the 1800 s cap so they cannot monopolise pass 1.

• consecutive_failures resets per range: the wedge-exit detector (>= 50 consecutive failures) is for "stuck on the same range"; pre-fix the counter persisted across ranges, so 50 small post-bisection ranges with one failure each falsely tripped the wedge mid-pass. Reset at every range boundary.

• Reverted inline 5× retry in patch: a brief experiment that was measurably harmful — each "2 s" SCSI timeout paid ~1.5 s kernel SCSI mid-layer error-escalation overhead, so 5× retry took ~17 s per LBA and triggered the per-range watchdog in 4 sectors. Restored READ_RECOVERY_TIMEOUT_MS = 60_000 (the v0.17.3 baseline; the kernel- auto-retry pattern is now provided by the /dev/sr0 fallback above).

Empirical results (a UHD disc, BU40N direct SATA)

• Pass 1: 94.6 MB recovered (28% of formerly-bad data, 33 sr0 fallback saves), 11 s of main-title content restored. Patch completed all 47 retryable ranges naturally (was wedging at range 1 of 47 in v0.17.3).
• Pass 2 cumulative: 95.2 MB (+0.6 MB; diminishing returns curve).
• Remaining ~233 MB on the test disc appears physically unrecoverable on this hardware (kernel auto-retry can't decode it either).

Behavioural notes

• The /dev/sr0 fallback is Linux only; macOS and Windows fall back to the existing single-shot SCSI behaviour. The fallback is gated to recovery=true reads (only fires from the patch path, not the sweep path) to avoid page-cache pressure during multi-GB sequential ripping.

0.17.0 (2026-05-04)

Code quality: unwrap safety, clippy compliance, test coverage

• Sweep() hot path: Replaced .err().unwrap() in disc/mod.rs:1553 with explicit pattern matching (match read_result { Ok(_) => unreachable!(), Err(e) => e }). The original unwrap was logically safe (in the else branch after is_ok() check) but pattern matching makes the invariant explicit and avoids panic risk if logic changes.
• cargo clippy --lib: Clean build with -D warnings across all targets. No lint failures.
• cargo test --lib: All 256 tests pass (0 failed, 1 ignored). Integration tests for sweep/patch/dev-null validate the multi-pass recovery pipeline end-to-end.

Patch pass algorithm fix

• patch_internal excludes Unreadable sectors from work list (only retries NonTrimmed/NonScraped/Unscraped). Previously attempted to retry already-failed sectors, wasting time and masking real progress.
• Exposes bytes_bad_in_title in patch results for accurate UI reporting of unrecoverable data vs. recovered data.

0.16.1 (2026-04-30)

Unified progress display for sweep and patch

• Patch passes now show bytes_good / disc_total as GB (grows when sectors are recovered), with pass progress % using work_done / work_total (always advances). Xs unreadable declines as data is recovered. Same UI for all pass types.

0.16.0 (2026-04-30)

IOKit registry-based drive enumeration, BSD name matching, reverse patch default

• shim_open_exclusive now matches the correct IOBDServices for the requested BSD name. Walks all IOBDServices entries in the IOKit registry, matches child IOMedia "BSD Name" property. Falls back to IOMedia parent walk, then first-match. Fixes multi-drive systems.
• shim_list_drives (new): registry-based drive enumeration. Reads IOBDServices "Device Characteristics" for vendor/model/firmware and child IOMedia for BSD name. Zero SCSI, zero exclusive access, zero unmounts. Fixes the enumeration blast that unmounted every disk on the system.
• list_drives() and find_drives() rewritten to use shim_list_drives. No longer iterates /dev/disk0..15 opening exclusive SCSI on each.
• patch_internal defaults reverse: true. Sweep jumps forward with escalating gaps, so NonTrimmed ranges have good data at their tail. Reverse patch hits good data first, converges on actual bad block boundaries.

0.15.1 (2026-04-30)

Fix damage-jump detection, fix dispatch covers_disc check

• Damage-jump tuning: DAMAGE_WINDOW=16, DAMAGE_THRESHOLD_PCT=12% (was 50/25%). Old params were too diluted by good reads between sparse failures; new params trigger on 2nd scattered failure.
• Disc::copy() dispatch: covers_disc now checks map.total_size() == disc_size (byte-for-byte, not approximate). Fixes false sweep dispatch when mapfile exists but doesn't cover full disc.
• Sweep resume: when dispatched from existing mapfile with NonTried, passes resume: true.

0.15.0 (2026-04-30)

Multipass dispatch rewrite, speed control on damage zone entry/exit

• CopyOptions { decrypt, multipass, progress, halt }. Disc::copy() auto-detects sweep vs patch from mapfile state: no mapfile → sweep, NonTried → sweep with resume, only NonTrimmed/NonScraped/Unreadable → patch, clean → no-op.
• Disc::mapfile_for(): /dev/null output → /tmp/<disc_name>.mapfile, otherwise mapfile_path_for(path).
• Speed control: damage zone entry → set_speed(0x0000) (minimum), 16 consecutive good reads → set_speed(0xFFFF) (maximum). Drive manages optimal speed in clean sections.
• SET CD SPEED SCSI command via SectorReader::set_speed() (default no-op, Drive impl sends SCSI).

0.14.0 (2026-04-30)

Damage-jump algorithm replaces probe, bridge degradation detection

• Damage-jump algorithm: when damage threshold exceeded in sliding window, jump ahead by 256×batch×multiplier sectors. Doubles multiplier on each jump. Zero-fills gap as NonTrimmed.
• ecc_sectors(): returns ECC block size per disc format (32 for UHD, 16 for BD, 16 for DVD).
• Bridge degradation detection: NOT READY with sense key 2/ASC 0x04/ASCQ 0x3E triggers 10s cooldown, up to 5 times before treating as bad sector.

0.13.43 (2026-04-29)

Pass 1 transport-failure recovery loop

• Transport failure (USB bridge crash) no longer kills the entire rip.
• Autorip re-discovers the drive after USB re-enumeration and resumes from the mapfile. Up to 10 attempts.
• Error::is_scsi_transport_failure() now matches DiscRead with status 0xFF (bridge crash) in addition to ScsiError.
• DriveSession tracks device_path for re-discovery.

0.13.42 (2026-04-29)

Fix: transport failure skips instead of aborting

• Transport failure (USB bridge crash) now skips the failed ECC block (marks NonTrimmed) and continues. 3 consecutive transport failures still abort the copy. Previously, a single transport failure killed the entire rip.

0.13.41 (2026-04-29)

Debug logging for sector-0 regression diagnosis

• Add debug logging to Drive::read and Disc::copy first reads.
• No functional changes.

0.13.40 (2026-04-28)

Pass 1 pure ECC-block sweep, transport-failure abort, mapfile-based recovery

This release reworks the sector-copy pipeline to handle unreliable USB-SATA bridges (notably the Initio INIC-1618L) that crash on MEDIUM ERROR retries.

Disc::copy() — Pass 1 (ECC-block sweep):

• Reads batch sectors (default 32 = 1 BD ECC block = 64 KB).
• Success → mark Finished. MEDIUM ERROR → zero-fill, mark NonTrimmed, advance.
• Transport failure (host_status=7) → abort immediately, return error.
• No single-sector reads, no retry, no state machine in Pass 1.

Disc::patch() — Pass 2+ (single-sector recovery):

• Reads NonTrimmed sectors one at a time with pause between failures.
• Succeeds → mark Finished. Fails → mark Unreadable (or leave for next pass).
• Multi-pass: caller runs patch repeatedly until 0 recovered.

CopyOptions simplified:

• Removed skip_forward, cautious_pause_ms, BPT1_EXIT_THRESHOLD.
• Fields: decrypt, resume, batch_sectors, skip_on_error, progress, halt.

Other fixes since 0.13.26:

• open() just opens the device — no side effects. drive_has_disc() is a standalone TUR, not a probe sequence.
• enumerate_sg_names() skips unreadable /sys type files.
• SCSI sense data preserved in DiscRead errors (status + key + ASC/ASCQ).
• MapStats splits bytes_pending into nontried / retryable.
• Wallclock rip budget — halt after max(disc_runtime_secs, 3600).
• Patch instrumentation: counters for reads_ok/err, writes_ok/err, finished/unreadable.
• as_encoded_bytes() replaces as_bytes() for portable OsStr handling.
• Removed inline retry/reset from Drive::read — orchestration layer handles recovery.
• Removed all internal references from public code.

0.13.26 (2026-04-27)

Extend DiscRead with SCSI status/sense for 30% wedge diagnostics

Error::DiscRead now carries status (SCSI status byte) and sense (ScsiSense with key/asc/ascq). Previously this info was discarded, showing only E6000: {sector}. Now shows:

• E6000: {sector} 0x{status}/0x{sense_key}/0x{asc}
• Enables recovery loop to distinguish recoverable errors from drive wedge
• Enables programmatic handling: if status == 0xFF { reset } else { retry }

Error display shows up to 5 fields

Display format changed from E{sector} to E{code}: sector status/key/asc.

0.13.25 (2026-04-27)

Drop dead Drive::device_path_owned()

The method was marked // NOTE: Debug aid — remove after fd issue is resolved and the fd issue closed in 0.13.6. Use device_path() (which returns &str) instead. Removing it clears a cargo clippy -- -D warnings red on Linux CI that the Mac toolchain doesn't catch.

Pre-commit gate uses CI's exact toolchain

A new pre-commit script runs cargo +1.86 fmt --check, cargo +1.86 clippy -- -D warnings, and cargo +1.86 test --tests across all 5 freemkv crates. Mirrors each repo's .github/workflows/ci.yml step-for-step. Use it as a pre-commit hook or run by hand before pushing — green here means green CI.

The Mac default toolchain is newer (1.94) and its clippy rejects slightly different sets of lints than 1.86 — running locally without pinning misses lints CI catches. The script forces 1.86 so drift between local and CI ends.

0.13.24 (2026-04-27)

MapStats: split bytes_pending into bytes_nontried + bytes_retryable

MapStats.bytes_pending aggregates NonTried (sectors Pass 1 hasn't reached) + NonTrimmed + NonScraped (sectors flagged for Pass 2-N retry). UIs that wanted a "MAYBE / will retry" bucket were stuck showing the entire unread disc as "Maybe" at pct=0.

v0.13.24 keeps bytes_pending for back-compat and adds two granular fields:

• bytes_nontried — Pass 1 hasn't read these yet
• bytes_retryable — NonTrimmed + NonScraped, Pass 2-N will retry

bytes_pending == bytes_nontried + bytes_retryable (invariant).

cargo fmt cleanup

Picks up the cargo fmt --check lint failure that's been red on main since v0.13.18 (long format-string layouts the local rustfmt folded differently from CI's runner).

0.13.23 (2026-04-27)

Stop discarding the drive's SCSI sense data

Through the entire 0.13.x line, every CHECK CONDITION reply from the drive (the standard way SCSI reports a sector failure) was being collapsed into a synthetic status=0xFF, sense_key=0 "transport wedge" sentinel and the real sense data was thrown away. Live tracing on the BU40N reading a UHD disc on 2026-04-27 confirmed it: the drive was returning host_status=0, driver_status=8, status=2, exec_elapsed_ms=1416 on every bad sector — a clean CHECK CONDITION carrying full sense data — and the library was misclassifying it as a wedge and bailing.

Root cause: scsi/linux.rs's wedge check was host_status != 0 || driver_status != 0. SG's DRIVER_SENSE bit (0x08) is set on every CHECK CONDITION reply just to flag "sense buffer is populated" — it's not a transport failure on its own. Pre-0.13.23 we conflated the two and silently lost every drive-reported error reason. macOS and Windows backends had the same shape: they extracted sense_key only, dropping ASC/ASCQ.

What 0.13.23 changes (API)

• Linux: mask DRIVER_SENSE before treating driver_status as a transport-layer failure. Real transport failures (host_status != 0 or any non-SENSE bit set) still synthesise the 0xFF sentinel.
• Error::ScsiError carries sense: Option<ScsiSense> instead of flat sense_key/asc/ascq. sense=None ⇔ transport failure (no SCSI status delivered). Some(ScsiSense {…}) ⇔ drive replied with sense data. Removes the 0xFF/sense_key=0 magic-number coupling.
• ScsiSense is a public type with predicate methods on it — is_marginal, is_medium_error, is_hardware_error, is_unit_attention, is_data_protect, is_not_ready, is_illegal_request, is_aborted_command. Callers route on the structured fields rather than raw key comparisons.
• Error::scsi_sense() / Error::is_scsi_transport_failure() / Error::is_marginal_read() convenience predicates on Error. is_marginal_read is the high-level "should Disc::copy engage hysteresis on this error?" check.
• SCSI protocol constants (SCSI_STATUS_GOOD, SCSI_STATUS_CHECK_CONDITION, SCSI_STATUS_TRANSPORT_FAILURE, SENSE_KEY_*) moved from error.rs to scsi/mod.rs where they belong alongside SCSI_INQUIRY, SCSI_READ_10, etc.
• macOS + Windows backends parse the full sense triple too. Same code path on every platform — a regression in parse_sense would surface on all three OSes simultaneously.
• parse_sense replaces parse_sense_key (returns the full triple). Inline sense-format tests (descriptor 0x72/0x73 vs fixed 0x70/0x71, short-buffer, VALID-bit masking, unknown response codes) now also exercise ASC/ASCQ extraction at the right offsets.

Disc::copy + Disc::patch sense-aware dispatch

Both passes now bail immediately when a read fails with a sense class that retry can't help (HARDWARE ERROR, DATA PROTECT, UNIT ATTENTION, NOT READY, ILLEGAL REQUEST, real transport failure, kernel IoError) rather than burning hysteresis cycles on a doomed loop. Marginal-read sense (MEDIUM ERROR, ABORTED COMMAND, RECOVERED ERROR, NO SENSE) engages hysteresis as before. New phase=bail trace event records the bail reason.

Disc::patch's wedged_threshold (50 consecutive failures) remains as defense-in-depth for chains of marginal failures, but a single non-marginal sense now short-circuits it.

Behavioural impact

For damaged-disc rips on the BU40N this unblocks v0.13.22's hysteresis: pre-fix, the misclassified "wedge" caused Disc::copy to exit before hysteresis could engage, so bytes_good froze at the bad zone. Post-fix the drive's CHECK CONDITION replies flow through the normal path → hysteresis drops to bpt=1 → marginal sectors are recovered or marked Unreadable. Calibration data (docs/audits/2026-04-26-bisect-on-fail-empirical-findings.md) shows ~86 % of marginal-region sectors recover at bpt=1 on this drive.

0.13.22 (2026-04-26)

Replace bisect-on-fail with hysteresis state machine (Block ↔ Single)

Live test on a UHD disc v0.13.21 showed bisect-on-fail recovered every recoverable sector, but spent ~30 sec per damaged 60-block (paying a ~5 sec kernel timeout at every bisection level). Each level descended log₂(60) ≈ 6 times on the failing branch.

Replaced with a two-state hysteresis machine in Disc::copy:

Block(batch):
  read(batch) ok    → write, advance, stay Block
  read(batch) fail  → switch to Single, retry SAME range at bpt=1

Single:
  read(1) ok    → write, consecutive_good++
                  if consecutive_good >= BPT1_EXIT_THRESHOLD:
                     switch to Block, reset counter
  read(1) fail  → mark NonTrimmed, consecutive_good = 0

BPT1_EXIT_THRESHOLD = 10_000 sectors (= 20 MB of clean data). Calibrated from the 2026-04-26 BU40N empirical run; tunable.

Per-block cost on a 60-sector damaged block with 1 bad sector:

• Bisect (v0.13.21): ~30 sec (5 s × 6 levels)
• Hysteresis (v0.13.22): ~10 sec (5 s bpt=batch fail + 59 × 1 ms good
  • 1 × 5 s bad)

Plus inside a damaged cluster spanning many 60-blocks, hysteresis pays the bpt=batch fail cost ONCE on entry; bisection paid it every 60 sectors. For the ~1248-sector boundary cluster that's ~21 fewer 5-sec waits = ~100 sec saved.

Telemetry: new phase=mode_change trace event with from, to, lba, consecutive_good. Replaces v0.13.21's phase=bisect. The v0.13.21 worklist DFS is gone — single iterative for s in 0..count on the failure path.

Test rename: test_disc_copy_bisect_recovers_via_single_sector_reads → test_disc_copy_hysteresis_recovers_via_single_sector_reads. Same synthetic BU40N-pattern reader; same 100% recovery expectation.

0.13.21 (2026-04-26)

Fix: Disc::copy bisect-on-fail (replaces skip-forward)

Empirical live-hardware testing on the LG BU40N (see the internal test-plan audit and run log) revealed that the drive often fails multi-sector READ commands in damaged regions but succeeds when asked one sector at a time. The old skip-forward strategy responded to multi-sector failures by jumping up to 1 % of the disc forward, marking everything in between as bad — losing clean territory sandwiched between bad sectors.

Disc::copy now bisects on read failure: split the failed block in half, retry each half, recurse down to single-sector reads. Sectors the drive can read individually are recovered in Pass 1; only sectors that fail at bpt=1 are marked NonTrimmed for the patch passes.

Empirical results on a UHD disc on the BU40N:

• Old algorithm: 25 GB read in Pass 1, then ~6 GB skip-forwarded; retry passes failed to recover most of the skipped zone.
• New algorithm: ~99 % of disc recovered in Pass 1; only the truly unreadable cluster (~14 % of a 2 MB hot zone) marked NonTrimmed.

Implementation: stack-based DFS in the inner read loop. log₂(batch) levels max — for the default 60-sector batch, 6 levels. Multi-pass machinery is untouched: Pass 2 .. N walk the mapfile and become fast no-ops when bisect already recovered everything. New integration test test_disc_copy_bisect_recovers_via_single_sector_reads validates the behavior against a synthetic BU40N-pattern reader.

Fix: READ_TIMEOUT_MS bumped 1.5 s → 10 s (caller-side)

The 0.13.20 SCSI rewrite gave the kernel mid-layer the ability to run its own ABORT/RESET escalation. But callers (Drive::read for the fast path) still passed timeout_ms=1500. Cold-start seek on the BU40N can take ~1.5 s, which means normal reads were being cancelled at the boundary, triggering the kernel mid-layer's escalation, which the Initio bridge couldn't drain — resulting in the firmware-level wedge that only physical replug recovers.

Live-hardware probe data:

• Sustained sequential read: 3–7 ms
• Cold-start seek + read: up to ~1500 ms
• Successful ECC recovery: 1.6–2.6 s
• Confirmed unreadable: 3.6–8.8 s (kernel timeout)

10 s is calibrated to cover every legitimate read with margin while still short-circuiting truly bad sectors before the kernel runs full LUN/BUS/HOST reset. READ_RECOVERY_TIMEOUT_MS (60 s) unchanged.

0.13.20 (2026-04-26)

Architecture: SCSI transport — sync blocking SG_IO

scsi/linux.rs rewritten from async `write/poll/read + 1.5 s timeout

• close-on-timeout in bg threadto a single synchronous blockingioctl(fd, SG_IO, &hdr). The old pattern abandoned slow-but-alive commands faster than the drive could drain its internal queue, deepening the BU40N wedge. Per the audit at internal SCSI-architecture research, no reference project (MakeMKV / sg_dd / ddrescue) does what we did — all use sync blocking SG_IO with 8-60 s timeouts and let the kernel's mid-layer (scsi_eh.rst`) run ABORT TASK / LUN RESET / BUS RESET / HOST RESET escalation internally.

What changed:

• SgIoTransport::execute() is one syscall now. Caller-supplied timeout_ms is honored by the kernel, which does its own ABORT/RESET escalation if the device times out.
• Errors check host_status and driver_status (both 0xFF-synthesised for the caller) in addition to status — transport-level failures no longer slip through as Ok.
• Sense-key parser handles both descriptor format (0x72/0x73, key at byte 1) and fixed format (0x70/0x71, key at byte 2).
• Deleted the fd_recovery: Arc<AtomicI32> field, the bg close+open thread, and the stale-fd swap dance. scsi/linux.rs shrank from ~720 to ~520 lines.
• Module doc rewritten to reflect the new architecture.

Architecture: parity strip on macOS + Windows

scsi/macos.rs and scsi/windows.rs had try_recover() — userspace handle-recovery on task failure. Same anti-pattern as the Linux fd-recovery dance, removed for the same reason: the kernel mid-layer already runs its own escalation. Errors bubble up directly.

Cleanups:

• MacScsiTransport: try_recover() deleted, bsd_name field deleted (was only used by try_recover), fail-fast device_iface guard deleted (no longer null'd mid-session).
• SptiTransport: try_recover() deleted, wide_path field deleted, INVALID_HANDLE guard deleted.

API cleanup: drop Drive::reset and find_drives

Two duplicates removed from the public surface:

• Drive::reset() — escalating recovery (STOP/START unit + eject + reinit). Per the audit, userspace shouldn't escalate; the kernel already does. Only one internal caller (wait_ready line 195), which now just keeps polling TUR for 60 iterations. No external consumer used it.
• pub fn find_drives() -> Vec<Drive> — opened N drives just to throw most away. Only caller was find_drive() itself, which now uses discover_drives() directly. No external consumer used it. For lightweight enumeration (UI sidebar etc.) use scsi::list_drives().

lib.rs re-export of find_drives removed.

0.13.19 (2026-04-26 — held, never released)

Held in development; folded into 0.13.20.

0.13.18 (2026-04-26)

Sync release — no functional changes

Bumped to satisfy the unified-versioning rule. Actual fix is in autorip (web.rs two-bar UI — separates per-pass and total progress bars + their own text rows so the rip dashboard is readable again).

0.13.17 (2026-04-26)

Sync release — no functional changes

Bumped to satisfy the unified-versioning rule. Actual fix is in autorip (hot-plug rescan in the drive poll loop — autorip now picks up unplug/replug events without a container restart).

0.13.16 (2026-04-26)

Architecture: single Progress trait + PassProgress struct

Pre-0.13.16 the rip API leaked internal mapfile concepts (pos, bytes_good, work_done, bytes_pending, Finished/NonTrimmed) into per-pass positional callbacks. Consumers reinvented the math each time, and the v0.13.15 UI bug surfaced exactly because of this — autorip's web JS computed progress_pct from bytes_good while the backend computed from pos, silent drift, frozen UI bar.

This release replaces both Disc::copy::on_progress and Disc::patch::on_progress Fn(u64, u64, u64) callbacks with a single Progress trait and PassProgress struct (new progress module).

pub struct PassProgress {
    pub kind: PassKind,            // Sweep | Trim {reverse} | Scrape {reverse} | Mux
    pub work_done: u64,
    pub work_total: u64,
    pub bytes_good_total: u64,
    pub bytes_total_disc: u64,
}

pub trait Progress {
    fn report(&self, p: &PassProgress);
}

impl<F: Fn(&PassProgress)> Progress for F { ... }

Both CopyOptions::on_progress and PatchOptions::on_progress are renamed to progress: Option<&dyn Progress>. Closure callers update trivially; struct callers gain a clean named-field shape with no positional-arg confusion.

PassKind carries the semantic (sweep vs trim vs scrape vs mux) so consumers can label phases without reinventing detection logic. The Mux variant is reserved for v0.13.17 when the mux pipeline emits progress; not yet emitted by libfreemkv code.

Disc::patch reports Trim {reverse} for retry passes with block_sectors >= 2 and Scrape {reverse} when block_sectors == 1 (the per-sector final pass). Direction comes through reverse: bool.

0.13.15 (2026-04-26)

Breaking: on_progress callback gains pos parameter

Both CopyOptions::on_progress and PatchOptions::on_progress now take Fn(bytes_good: u64, pos: u64, total_bytes: u64). The new pos parameter is the current sweep / retry position. Pass 1 callers should display pos / total_bytes for the "% swept" UI bar — bytes_good only counts clean reads (Finished sectors) and freezes during skip-forward bad zones, which made every previous version's UI look hung at the bad-zone boundary. This was the v0.13.9 stall-guard origin bug.

Live trace from v0.13.14: Pass 1 hit a bad zone at 24 GB and appeared "stuck" for 14 minutes per autorip's UI (bytes_good = 23.97 GB unchanged). Disc trace events showed pos actually advanced from 25.8 GB to 70 GB during that window — Pass 1 was 83 % through the disc, marking the post-bad-zone NonTrimmed via skip-forward exactly as designed. The display lied. Now consumers can show the truth.

Feature: PatchOptions::reverse for reverse-direction retry passes

When set, Disc::patch walks bad ranges from highest LBA to lowest, and within each range reads sectors back-to-front. Hypothesis (per the live v0.13.14 test): drives that wedge after a forward read of a bad sector read fine when approached from end-of-disc backward — most of the post-bad-zone NonTrimmed range is actually clean data the drive could have read on Pass 1 had it not been wedged. autorip alternates F/R across retry passes (Pass 2 = reverse half-batch, Pass 3 = forward quarter-batch, ...).

Feature: PatchOptions::wedged_threshold early-exit

When > 0, Disc::patch exits early if it sees this many consecutive read failures with zero successful reads in the same pass. Saves the wallclock budget for productive grinding when the drive has clearly wedged on the bad zone for this pass — a future pass with a different direction or block size may still recover. Reported via new PatchResult::wedged_exit: bool.

Trace: patch_start and patch_done events

freemkv::disc target now emits patch_start (block_sectors, recovery, reverse, wedged_threshold, num_ranges) and patch_done (blocks_attempted, blocks_read_ok, blocks_read_failed, wedged_exit, halted, bytes_recovered) at Disc::patch boundaries.

0.13.14 (2026-04-25)

Sync release — no functional changes in libfreemkv

Bumped solely to satisfy the unified-versioning rule. The actual fix in this release is in autorip: the tracing subscriber now enables freemkv::scsi=trace,freemkv::disc=trace so the v0.13.13 instrumentation events actually surface in /api/debug. Without that filter override the trace events were silently dropped by the default libfreemkv=warn rule.

0.13.13 (2026-04-25)

Telemetry: instrument the rip pipeline for in-flight diagnosis

v0.13.12 shipped Fix 1+2+4 + cross-platform parity but a live test showed Pass 1 sat for 14 minutes with bytes_good=0 while the inner loop appeared to iterate (the throttled on_progress log fired every 78s). The async fd_recovery design at §7 said each execute() call should bound at ~1.5 s on poll timeout, with subsequent calls returning DeviceNotFound in microseconds until recovery completes. Observed reality contradicts that: each iteration takes ~60 s, not microseconds. Without trace-level telemetry at the SCSI + Disc::copy boundaries we can't diagnose where the time goes.

This release adds the telemetry. No behavior change; instrumentation only.

• New dep: tracing = "0.1". Per project docs, debug/trace logging is permitted in libfreemkv (the no-English rule applies to errors, not telemetry). Consumers (autorip) wire a tracing subscriber and pipe events into the JSONL debug log automatically.
• SgIoTransport::execute (Linux): trace events at every state transition (entry, recovery_swap_ok, recovery_pending, write_ok / write_err, poll_done, timeout_spawn_recovery, scsi_err, read_err, ok). Each event includes the opcode and elapsed timing. The bg recovery thread also traces close_ms + open_ms so we can see if the kernel is hanging close+open.
• Disc::copy: trace events at copy_start, outer_loop, region_enter, every 100 inner-loop iterations (iter_progress with pos / region_end / skip_size / bytes_good / read_ok_count / read_err_count / last_read_ms / copy_elapsed_ms), and copy_done.
• All trace events use target strings freemkv::scsi and freemkv::disc so consumers can filter by subsystem.

What this enables

• A live rip will now produce a SCSI event stream visible at /api/debug?n=N&q=freemkv::scsi. We can finally answer: "is the inner loop iterating slowly because each call is slow, or fast with the bg thread blocked?"
• bg_recovery_done events with close_ms / open_ms reveal whether the kernel really takes 60 s for close+open on a wedged Initio bridge.

0.13.12 (2026-04-25)

Fix: delete stall guard from Disc::copy (RIP_DESIGN.md §6 Fix 1)

The v0.13.9 stall guard at disc/mod.rs exited Pass 1 early when bytes_good was flat for stall_secs (default 120s). This violated the ddrescue model: Pass 1 must sweep end-to-end, marking failed reads NonTrimmed for Pass 2 retry. The guard caused Pass 1 to bail at 30% on a UHD disc with 56 GB still NonTried, leaving Pass 2 nothing useful to do.

• Deleted the stall-guard state vars and the if cur_good != ... break 'outer; block.
• Deleted CopyOptions::stall_secs field — no longer wired.
• Replaced the broken regression test test_disc_copy_stall_detection_triggers_skip_forward with test_disc_copy_completes_full_disc_with_failing_reader (asserts Pass 1 walks to end-of-disc with everything NonTrimmed when reads keep failing) and added test_disc_copy_halts_promptly_on_failing_reader (halt flag honored within 2s mid-skip-forward).

Fix: async SCSI transport recovery (RIP_DESIGN.md §6 Fix 2 / §7)

SgIoTransport::execute (Linux) previously did close-in-background + synchronous open-on-main-thread on poll timeout. The kernel serialized the main-thread open() against the in-flight close() of the same /dev/sg*, blocking the rip thread up to ~60s per timeout.

• Added fd_recovery: Arc<AtomicI32> field. On poll timeout, both close(old_fd) AND open(new_fd) run in a background thread; the new fd is published to fd_recovery. Returns Err immediately. Main thread is never blocked beyond the poll() budget (~1.5 s).
• Top of execute(): if self.fd < 0, swap from fd_recovery. If recovery is also pending, return DeviceNotFound and let the caller's retry loop come back later.
• Drop drains any pending fd_recovery so the fd doesn't leak.
• Stripped the v0.13.9 stall-guard narrative comment that justified the deleted behavior.

Fix: cross-platform SCSI parity — Windows + macOS recovery (RIP_DESIGN.md §15.1)

Per the platform parity rule (no stubs), Windows and macOS now have the same observable recovery contract as Linux:

• SptiTransport (Windows): added try_recover() that calls CloseHandle + CreateFileW synchronously after a failed DeviceIoControl. Stripped English error string ("run as administrator") from open(). Fixed the ms→s timeout truncation (1500ms now rounds up to 2s, was 1s).
• MacScsiTransport (macOS): added try_recover() that releases the IOKit interface (RELEASE_EXCLUSIVE + com_release) and re-acquires via the new acquire_device_iface() helper. Stores bsd_name so recovery can re-call find_scsi_service.
• All three platforms: top of execute() returns DeviceNotFound immediately if a prior try_recover() left the transport in an invalid state. Drop guards null'd-out interfaces.
• Send is auto-derived on all three (i32 fd / isize HANDLE / IOKit interface ref are Send-safe); explicit comments document the intentional implicit Send and the absence of Sync.

Fix: instrument Disc::patch — diagnostic counters (RIP_DESIGN.md §6 Fix 4)

PatchResult now reports blocks_attempted, blocks_read_ok, blocks_read_failed. Pass 2's "100 minutes recovered 0 bytes" mystery (a UHD disc) becomes diagnosable from these counters: distinguish "drive returned Ok but write/record dropped data" from "every read was Err for the entire range" without instrumenting from outside the lib.

Fix: honor PatchOptions::full_recovery

The field was previously read into let _ = opts.full_recovery; and ignored — read_sectors(..., true) was hardcoded. Now routed to read_sectors(..., opts.full_recovery). Behavior unchanged for default callers (which pass true).

Doc: CopyOptions::batch_sectors accuracy

Doc comment said "Defaults to 32 sectors (64 KB)". Updated to describe the actual production path: callers should resolve via detect_max_batch_sectors(device_path) (kernel-reported sysfs value, typically 60 sectors / ~120 KB on the BU40N). The 32-sector internal fallback is only reached when batch_sectors=None AND skip_forward=true.

0.13.11 (2026-04-25)

Fix: revert SgIoTransport timeout path to keep transport alive

v0.13.10 changed SgIoTransport::execute to set fd = -1 on a poll timeout (no reopen on the main thread, since that would serialize against the spawned close()). The intent was to escape the 60-s blocking reopen.

The cost was too high: a single transient poll timeout permanently killed the transport. Live test on a UHD disc (post-replug):

• Pass 1 ran for 45 ms then returned with 0 GB good and 80 GB pending.
• The first SCSI READ timed out, fd went to -1, every subsequent read returned DeviceNotFound instantly, Disc::copy raced through the entire disc skip-forwarding in milliseconds.
• Pass 2 inherited the dead Drive and was equally useless.

Revert: spawn close + reopen on main thread (the v0.13.5/8 behavior). Yes the main-thread open() may block up to ~60 s while the kernel completes the abandoned command — but the v0.13.9 Disc::copy stall guard already caps catastrophic stalls at 120 s of bytes_good non-advance. Net: per-timeout cost is ~60 s, but Pass 1 cleanly bails out within 120 s of any wedge, and Pass 2 has a working Drive to retry NonTrimmed ranges with recovery=true + 30 s timeouts.

The integration test for the stall guard (test_disc_copy_stall_detection_triggers_skip_forward) continues to pass — the guard fires regardless of which transport-recovery strategy is in play.

0.13.10 (2026-04-25)

Version sync — no functional changes

Sync bump for the autorip-side fix in 0.13.10 (Pass 1 batch reporting).

0.13.9 (2026-04-25)

Fix: Disc::copy silent stall + SgIoTransport reopen-after-timeout serialization

Two correlated fixes for a hang observed live on the LG BU40N during a v0.13.8 rip of a UHD disc. At ~30 % progress through Pass 1 (disc → ISO), bytes_good froze for 10+ minutes with errs=0, no error surfaced, drive not wedged.

Root cause: SgIoTransport::execute (linux.rs) attempted to recover from a poll() timeout by spawning a background close() of the old fd and opening a fresh /dev/sg* fd on the main thread. On Linux, opening the SAME device while a prior fd is mid-close serializes via the kernel's per-device state lock — so the fresh open() blocks for as long as the close does (until the kernel completes the in-flight CDB). This undid the userspace 1.5 s timeout: each timed-out read added 60+ s to the next iteration. From Disc::copy's perspective, reads kept returning Err slowly, the skip-forward path advanced pos but never bytes_good.

Fixes:

• SgIoTransport::execute no longer reopens on timeout. Spawns the close, sets self.fd = -1, returns Err immediately. Subsequent calls fail with DeviceNotFound (already gated at line 248). Caller (Drive) is invalidated until reopened. Pass 2's Disc::patch would need a fresh Drive; that's a v0.14 follow-up.
• Disc::copy adds a stall guard. New CopyOptions::stall_secs: Option<u64> (default 120 s). If bytes_good doesn't advance for the threshold, breaks the outer loop with complete: false, bytes_pending > 0 so the caller's retry path picks up.

Tests: new test_disc_copy_stall_detection_triggers_skip_forward in tests/integration_progress_and_halt.rs proves the guard fires within the configured threshold.

Other:

• Cosmetic: warning text "rip thread did not drain within 35s" updated to 60s (matches the v0.13.8 timeout bump).

0.13.8 (2026-04-25)

Version sync — no functional changes

Sync bump for the ecosystem. 0.13.8 carries autorip-side fixes: post-stop "error" leak (halt-aware Err handling in Pass 1/2+), 60 s drain timeout, and a structural spawn_rip_thread helper.

0.13.7 (2026-04-25)

Version sync — no functional changes

Sync bump for the ecosystem. All four freemkv crates (libfreemkv, freemkv CLI, bdemu, autorip) always share a version number; 0.13.7 carries an autorip-side fix (HTTP-spawned rip/scan threads now register for stop-drain).

0.13.6 (2026-04-25)

Inline retry/reset stripped from Drive::read; BytesRead now emitted

Two related changes that close the loop on the BU40N wedge work from 0.13.1–0.13.4 and on the long-standing autorip "0 KB/s, 0%" UI bug.

Drive::read is now single-shot. The phase 1 / 2 / 3 retry loop (reset → reopen → repeat) inside Drive::read is gone (~80 lines deleted). recovery=true only bumps the per-CDB timeout to 30 s; recovery=false keeps the 1.5 s timeout. On a failed read the function returns Err(DiscRead) immediately. Per the BU40N post-mortem, every USB / SCSI reset path tested in 0.13.1–0.13.3 resets the bridge but not the drive firmware, and the inline reset+reopen was the wedge primitive itself — issuing it from inside Drive::read produced multi-minute hangs and made the wedge class harder to surface to the user. The correct retry layer is Disc::patch's outer multi-pass loop, which is unaffected. A stuck drive now surfaces as a clean DiscRead to the caller, who can prompt physical replug.

SCSI reset surface trimmed. SgIoTransport::reset (Linux) drops the SG_SCSI_RESET ioctl and the STOP / START UNIT escalation; it keeps the kernel SG_IO state flush plus ALLOW MEDIUM REMOVAL. MacScsiTransport::reset is removed entirely (was open + drop + sleep, no SCSI). The top-level scsi::reset / scsi::reset_with_timeout / scsi::reset_blocking family is removed — no callers remain after the Drive::read strip.

EventKind::BytesRead now emitted. The variant was declared in 0.13.0 but never fired. DiscStream::fill_extents now emits BytesRead { bytes_read_total, total_extents_bytes } after every successful sector read, so consumers in direct (no-mapfile) mode can drive a real-time progress bar without polling output.bytes_written. Multi-pass mode continues to use Disc::copy's on_progress callback unchanged. Drives the autorip per-device live progress UI.

Drive::checked_sleep is removed (only used by the recovery loop); Drive::sleep_until_halted is #[cfg(test)]-only; Drive::emit is retained because BytesRead uses it.

Tests

• New tests/integration_progress_and_halt.rs (5 tests): BytesRead emission, Disc::copy on_progress regression guard, halt aborts copy, Drop safety, FileSectorReader round-trip.
• 233 unit tests + 5 integration tests pass.

Net diff

~80 lines deleted, ~20 added.

Version sync

0.13.6 ecosystem release (libfreemkv + freemkv + bdemu + autorip all on 0.13.6).

0.13.5 (2026-04-25)

Version sync — no functional changes

Sync bump for the ecosystem. All four freemkv crates (libfreemkv, freemkv CLI, bdemu, autorip) always share a version number; 0.13.5 carries autorip-side fixes (stop-is-reset, startup staging sweep).

0.13.4 (2026-04-25)

Wedge recovery rolled back + sysfs identity fallback

What changed. The in-library USB / SCSI wedge-recovery escalation added in 0.13.1 – 0.13.3 has been removed. drive_has_disc now returns the raw TUR result (or the 0xFF poll-timeout wedge error) directly to the caller. scsi::usb_reset() / usb_reset_with_timeout() / DEFAULT_USB_RESET_TIMEOUT_SECS and the per-platform SgIoTransport::usb_reset / MacScsiTransport::usb_reset / SptiTransport::usb_reset are gone. All three platform backends pass transport errors through verbatim, keeping the public list_drives + drive_has_disc contract symmetric (Linux / macOS / Windows).

Why. Production testing against the LG BU40N USB BD-RE (the drive that drove the whole 0.13.1–0.13.3 recovery push) showed:

• SG_SCSI_RESET, STOP UNIT + START UNIT, and USBDEVFS_RESET all succeed at the USB transport layer (kernel logs usb 3-2: reset high-speed USB device, device re-authorises).
• But the drive firmware below the USB bridge stays locked: no LUN enumerates on the fresh scsi_host, TUR never succeeds, /dev/sg* never reappears.
• Also tried (outside the lib): /sys/bus/usb/devices/<port>/authorized toggle, usb-storage driver unbind/rebind, forced SCSI host rescan. All same outcome.

Only physical unplug-replug (or host reboot) clears this wedge class. The library was logging 2-minute-per-tick escalation cycles for nothing, and consumers had no way to surface "drive needs physical intervention" to users because the escalation was masking the real failure. Upper layers (autorip, CLI) now see the wedge error directly and prompt the user.

A breadcrumb in scsi/linux.rs::drive_has_disc catalogues every recovery method tried and points to git tag v0.13.3 for the full implementation, in case a future hardware class is found where USB-layer recovery actually works.

New: sysfs-cached identity fallback (Linux). list_drives now populates empty INQUIRY vendor/model/firmware fields from /sys/class/scsi_generic/sgN/device/{vendor,model,rev} — the kernel runs its own INQUIRY at device probe time and stashes the answer there, so even a mid-wedge INQUIRY still yields the UI a human-readable identity. The drive surface on screen doesn't suddenly go blank the moment the drive firmware locks up.

0.13.3 (2026-04-24)

Bug fix — drive_has_disc wedge recovery was dead code for TUR errors

The wedge-signature predicate introduced in 0.13.2 gated on opcode == SCSI_INQUIRY (0x12) — a holdover from when enumerate-time INQUIRY was the only path wedges surfaced on. drive_has_disc issues TEST UNIT READY (0x00), so its wedge errors (E4000: 0x00/0xff/0x00) never matched the predicate and the SCSI-reset + USB-reset escalation never fired. Production result: the BU40N USB BD-RE stayed wedged indefinitely, with autorip logging recovery exhausted on the raw pass-through error while no recovery had actually been attempted.

Fix: drop the opcode constraint. Status byte 0xFF is synthesised by our own execute() path when poll() on the SG fd times out; it's the ground-truth wedge marker regardless of which opcode was in flight. Doc comments on is_wedge_signature and WEDGE_STATUS_BYTE updated accordingly. Linux-only — macOS / Windows use sense-key-based wedge detection and are unaffected.

0.13.2 (2026-04-24)

Public discovery + presence APIs; SCSI/USB primitives no longer

exposed to consumer crates

The autorip / freemkv-CLI side of the ecosystem was reimplementing hardware discovery (sysfs walking, SCSI type-5 filtering, sg-path construction) and SCSI recovery primitives in their own crates — a direct violation of the architectural rule that ALL hardware-aware code lives in libfreemkv. 0.13.2 closes that gap with two cheap public probes that absorb everything consumers were doing themselves, plus visibility tightening to make future violations a compile error.

New public APIs

• pub struct DriveInfo { path, vendor, model, firmware } — a single enumerated optical drive's identity. Returned by list_drives(), populated from a single SCSI INQUIRY at enumeration time. No firmware reset, no init.
• pub fn list_drives() -> Vec<DriveInfo> — one-shot enumeration across Linux/macOS/Windows. Linux walks /sys/class/scsi_generic/ with the SCSI type-5 filter and /dev/sg0..15 fallback; macOS walks /dev/disk0..15 with the INQUIRY peripheral-type-5 filter; Windows iterates CdRom0..15. Cheap (~10 ms / drive); cache the result and refresh on udev events.
• pub fn drive_has_disc(path: &Path) -> Result<bool> — single TEST UNIT READY. Returns Ok(true) when ready / Ok(false) on sense-key 2 ("medium not present") / Err only after recovery has been exhausted. Internal wedge recovery is hidden from callers — when the kernel returns the wedge-signature pattern (status 0xFF, no sense), this function transparently escalates: SCSI bus reset → if still wedged → USB device reset → retry TUR. Consumers never see the escalation.

USB-layer reset, multi-platform

USBDEVFS_RESET (Linux) is the only thing that recovers a kernel- level USB Mass Storage wedge — software equivalent of unplug-replug. Now wired for all three OSes:

• Linux: USBDEVFS_RESET ioctl on /dev/bus/usb/BBB/DDD. Resolves sg → USB device via sysfs walk (busnum/devnum parents).
• macOS: IOUSBDeviceInterface::ResetDevice(). Walks IORegistry parents from the SCSI service to the USB device, queries the IOKit USB plugin, calls ResetDevice.
• Windows: existing IOCTL_STORAGE_RESET_DEVICE covers both SCSI and USB layers via storport, so usb_reset returns DeviceNotFound by design — the recovery escalation in drive_has_disc falls through cleanly. (See the windows::usb_reset doc comment for why a separate cycle-port IOCTL isn't needed on Windows.)

All wrapped in a thread + mpsc::recv_timeout so a kernel ioctl that hangs forever can't lock up the caller (the inner thread leaks one OS thread per hard wedge — acceptable for a daemon that recovers vs. one that wedges the whole poll loop).

Visibility tightening (architectural enforcement)

These were pub in 0.13.1; consumer crates could (and did) call them directly, leaking SCSI knowledge across the lib boundary:

• scsi::reset → pub(crate)
• scsi::reset_with_timeout → pub(crate)
• scsi::usb_reset → pub(crate)
• scsi::usb_reset_with_timeout → pub(crate)
• DEFAULT_RESET_TIMEOUT_SECS / DEFAULT_USB_RESET_TIMEOUT_SECS → pub(crate)

Consumers now reach recovery exclusively through drive_has_disc, which folds the escalation in. Compile-time guarantee that no future autorip/CLI/bdemu commit can reintroduce direct SCSI access.

Why this design

Drive::open(path) runs a ~2 s firmware-reset preamble + identify sequence; suitable for ripping but wasteful for a poll loop probing "is there a disc?". Pre-0.13.2 autorip called Drive::open 4 × every 5 s = ~17 000 speculative SCSI sessions/day, hammering the drives between actual rips. The wedge in production at 23:51 UTC was triggered by exactly this hot-loop pattern. With drive_has_disc, the same poll cadence costs ~50 ms / drive (one TUR) — 40× cheaper and side-effect-free on a healthy drive.

Tests

• 233 lib tests pass (no change in count; APIs covered indirectly via the existing transport tests + a new device_key test on the autorip side).
• cargo clippy --all-targets -D warnings clean across Linux/macOS.

0.13.1 (2026-04-24)

scsi::reset() now has a hard wallclock timeout

Production incident on a wedged BU40N USB drive: autorip's poll loop called scsi::reset() and the call hung for 60+ seconds before the operator manually intervened. Root cause: the Linux SG_SCSI_RESET ioctl can block indefinitely when the kernel SCSI subsystem is waiting on a bus-wedged device that will never ack — there's no kernel-side timeout on this ioctl. Without an outer wallclock bound the caller's thread is stuck in the kernel until the device unwedges (which, for a permanently-dead USB target, may be never).

scsi::reset() is now a wrapper that runs the platform-specific reset on a detached worker thread and bounds the caller's wait via mpsc::recv_timeout(DEFAULT_RESET_TIMEOUT_SECS) (30 s). Returns DeviceResetFailed on timeout. The worker thread keeps running until the kernel eventually unblocks (we can't cancel a Linux ioctl from userspace) — this leaks one OS thread per hard wedge, an acceptable cost for a daemon that recovers vs. one that hangs.

• New pub const DEFAULT_RESET_TIMEOUT_SECS: u64 = 30;
• New pub fn reset_with_timeout(device, Duration) -> Result<()> for callers that want a different bound.
• Existing pub fn reset(device) -> Result<()> keeps the same signature; behaviour change is the timeout, not the API.

Follow-up flagged

SG_SCSI_RESET only resets at the SCSI layer. For USB-attached drives (the BU40N case), the wedge is often in the USB Mass Storage layer below SCSI — SG_SCSI_RESET doesn't help. The proper escalation is USBDEVFS_RESET (the usbreset.c ioctl), which re-enumerates the device at the USB layer. Tracked for 0.13.2: a scsi::usb_reset(path) that resolves sg → USB device and issues USBDEVFS_RESET. That would have recovered tonight's BU40N without operator intervention.

0.13.0 (2026-04-24)

Zero English in library — typed variants for every error path

Audit pass against the project docs rule (no English text in library code). Found nine call sites that violated the contract by stuffing English into io::Error::new(kind, "…") or by abusing Error::DeviceNotFound { path } as a free-form description field. Each is now a typed variant with structured fields; the CLI / autorip translates to localized text.

New Error variants and codes:

• ScsiInterfaceUnavailable { path } — E1004 (macOS SCSITaskDeviceInterface couldn't be obtained)
• DeviceLocked { path, kr } — E1005 (replaces an English "exclusive access denied. Try: diskutil unmountDisk" message)
• IoKitPluginFailed { path, kr } — E1006
• UnsupportedPlatform { target } — E2003 (built on an OS without an SCSI backend)
• PlatformNotImplemented { platform } — E2004 (replaces the product_revision: "Renesas not yet implemented" string-stuffing in drive::mod)
• MapfileInvalid { kind } — E6011 (ddrescue mapfile parse, with a stable &'static str kind: "status_char" or "hex")
• DiscUrlNotDirect — E9009 (replaces the full English sentence "Use Drive::open() + Disc::scan() + DiscStream::new() for disc sources" that mux::input(disc://…) returned to callers)

Migrated call sites:

• mux/resolve.rs — disc URL → DiscUrlNotDirect; the four format!("m2ts://…") / format!("mkv://…") IO error wraps now propagate the inner io::Error unchanged (the URL-prefix wrap added no semantic information).
• mux/iso.rs — same format!("iso://…") wrap dropped.
• sector.rs — image-too-large now uses the existing IsoTooLarge variant instead of format!("…image too large, max ~8 TB").
• disc/mapfile.rs — bad-status-char and bad-hex parser errors now use MapfileInvalid { kind }.
• scsi/mod.rs — unsupported-platform path uses UnsupportedPlatform.
• scsi/macos.rs — IOKit plugin failure → IoKitPluginFailed, SCSITaskDeviceInterface missing → ScsiInterfaceUnavailable, exclusive-access denied → DeviceLocked with the IOReturn code as a structured kr field. The find_scsi_service four-stage failure path is now a single DeviceNotFound { path } (none of the prior per-stage English descriptions were user-actionable individually).
• drive/mod.rs — Renesas platform → PlatformNotImplemented.

Stripped English from labels module

labels::apply() previously pushed "Commentary", "Descriptive Audio", "Score", "IME", and " (Secondary)" directly into AudioStream.label. Those English strings then leaked into MKV titles and into autorip's UI. The data was already structured upstream (LabelPurpose enum on StreamLabel); the lib was downcasting it for the caller's convenience.

• AudioStream gains purpose: LabelPurpose (re-exported from crate::disc next to the struct, alongside LabelQualifier).
• SubtitleStream gains qualifier: LabelQualifier.
• apply() writes structured fields, never English. label keeps codec-formatting only ("Dolby TrueHD 5.1").
• generate_audio_label drops the " (Secondary)" suffix; secondary is already a bool field, callers render it.
• generate_video_label drops the "Secondary Video" fallback. "Dolby Vision EL" kept (brand identifier, not translatable).

API hygiene

• mux module visibility tightened. pub mod ebml, m2ts, mkv, network, null, ps, stdio, ts, tsmux are now pub(crate). Their types are still re-exported from lib.rs — the modules themselves were leaking low-level EBML primitives, TS muxer internals, and network/stdio implementations that no external caller used. mux::codec, mux::disc, mux::iso, mux::resolve, and mux::meta stay public (genuine APIs).
• Stream trait rustdoc. The keystone PES Stream trait (in pes.rs) had per-method docs but no trait-level doc. Now explains read-vs-write split, error contracts, the role of info() / codec_private() / headers_ready().
• lib.rs re-export sections. Eight grouped sections with a paragraph each (Drive lifecycle, Errors, Decryption, Disc structure, Streams, Lower-level surfaces) so cargo doc tells callers when to reach for what.
• Dropped ScanOptions::with_keydb(). The _with_X constructor pattern was banned by project docs (one method per action). Use the struct literal: ScanOptions { keydb_path: Some(p.into()) }. Five external call sites (autorip ×3, freemkv CLI ×3) and three test fixtures migrated.
• pid_index allocation documented. The TsDemuxer::new flat lookup table was flagged by audit as "unbounded for adversarial PIDs"; on closer reading it's bounded by u16::MAX × 2 bytes ≈ 128 KB. Doc comment now states the bound explicitly so future contributors don't re-flag it.

Dead-code sweep

Pre-PES-rewrite leftovers that were pub but unreachable:

• Deleted mux/lookahead.rs entirely (orphan file — never had a mod declaration; only used by its own tests).
• Deleted mux/tsreader.rs (TsDemuxReader struct + four methods, used nowhere).
• Deleted mux::ebml::write_int, read_vint, SEEK_HEAD, SEEK, SEEK_ID, SEEK_POSITION (unused).
• Deleted mux::ts::scan_first_pts, scan_last_pts, scan_duration, SCAN_HEAD_SIZE, SCAN_TAIL_SIZE, take_remainder, set_remainder (unused since the v0.10 PES rewrite).
• Deleted MkvMuxer::codec_private_slots / codec_private_filled fields and fill_codec_private method — deferred-codecPrivate path was never exercised once codec_privates flowed through DiscTitle.

cargo clippy --all-targets -- -D warnings is clean.

Tests

• New error::tests module — code distinctness, Display has no English words, io::ErrorKind mapping for every new variant.
• 233 lib tests (was 230), all green.

Breaking changes

Source-compatible for callers who use Error opaquely (handle Result<T, Error> and error.code() only). The following are breaking:

• ScanOptions::with_keydb() removed — use struct literal.
• mux::ebml, mux::mkv, mux::ts, etc. modules no longer accessible externally — use the re-exported types from the crate root instead.
• AudioStream and SubtitleStream gained required fields (purpose, qualifier). Construction-by-struct-literal must include them.
• Error::UnsupportedDrive { product_revision: "Renesas not yet implemented" } no longer produced — match PlatformNotImplemented.

Magic-number policy

Per the v0.13 audit directive: new code in this release uses named documented constants (e.g. POLL_INTERVAL_SECS in autorip, the wedge signature literals in ripper.rs, the BD_TS_PID_SPACE floor in ts.rs's table allocation). A comprehensive retrofit of pre-existing magic numbers across the older codebase is queued as follow-up work for 0.13.1+ — too large to absorb into this release without scope creep.

0.12.0 (2026-04-24)

Rust 2024 edition migration

• Bumped edition = "2024". Required code changes:
  • FFI declarations in src/scsi/macos.rs wrapped in unsafe extern "C" { … } per the 2024 FFI safety rules.
  • unsafe_op_in_unsafe_fn lint: vtable_fn() body now has an explicit unsafe { … } block rather than relying on implicit unsafe of the containing unsafe fn.
  • Match-ergonomics: removed redundant ref/ref mut bindings in mux/meta.rs, mux/mkvstream.rs, mux/network.rs, mux/stdio.rs — 2024 tightens "cannot explicitly borrow within an implicitly-borrowing pattern."
• No behavior change. MSRV stays at 1.86.

Minor / version sync

• Part of the 0.12.0 ecosystem release. The autorip-side fixes (progress regressions, UI redesign, regression-guard tests) drove the minor bump.

0.11.22 (2026-04-24)

Version sync — no functional changes

Part of the 0.11.22 ecosystem release. autorip 0.11.22 ships full multi-pass UI (bad-range viz, live mapfile stats, Recovery settings); the library API is unchanged from 0.11.21.

0.11.21 (2026-04-24)

Multi-pass rip architecture — disc → ISO → patch → ISO

New primitives for two-stage rip: fast forward pass with zero-fill on failures, then targeted retries of bad ranges. Keeps the library API stream-based; the multi-pass model lives entirely in caller-orchestrated function composition.

• New Disc::copy(reader, path, &CopyOptions) replaces the positional-arg version. Always produces a ddrescue-format mapfile at path + ".mapfile" as a side-effect. With skip_on_error=true + skip_forward=true, does ddrescue-style fast sweep: 64 KB block reads, exponential skip-forward (256 KB → cap at 1% of disc) on failure, zero-fill bad blocks, record ranges in the mapfile. With defaults (both false), matches pre-0.11.21 behavior — uses drive-level recovery, aborts on bad sector. Mapfile is produced either way.
• New Disc::patch(reader, path, &PatchOptions) — idempotent retry pass. Reads the mapfile, re-reads every non-+ range with full drive recovery enabled, writes successful bytes back into the ISO at exact offsets, updates mapfile. Call N times for N retry attempts.
• New disc::mapfile module — ddrescue-compatible plain-text format. Crash-safe (flushes on every record()), greppable, human-editable, tool-interoperable. Status chars match ddrescue: ? non-tried · * non-trimmed · / non-scraped · - unreadable · + finished.
• Re-exports: FileSectorReader from the crate root for ISO readers.

Breaking changes

• Disc::copy's signature changes from positional args (decrypt, resume, batch, on_progress) to CopyOptions. Previous callers must migrate. freemkv CLI updated in lockstep.

Version sync

• Part of the 0.11.21 ecosystem release (libfreemkv + freemkv + bdemu + autorip all on 0.11.21).

0.11.18 (2026-04-24)

DiscStream halt flag — Stop works during dense bad-sector regions

DiscStream::fill_extents loops internally when the demuxer hasn't accumulated enough data to emit a PES frame — during a dense bad-sector run, that loop can spend many minutes shrinking batch sizes and zero-filling sectors without ever returning to the outer read() call. Without an internal halt check, the caller's Stop request goes unserviced until the demuxer eventually emits a frame, which may be very far away.

• DiscStream::set_halt(Arc<AtomicBool>) — share a halt flag with the stream. Typically wired to Drive::halt_flag() so Stop propagates across both the drive's recovery phases and the stream's sector processing.
• fill_extents() checks the halt flag at the top of every retry iteration (before each attempt at every size level). Raising the flag aborts within one read round-trip — at most the current SCSI command's timeout.
• Returns Err(Error::Halted) (E6010) so the outer rip pipeline terminates cleanly.

No behavior change for callers that don't call set_halt. Unblocks the architectural fix for the "Stop doesn't stop" bug observed on a damaged UHD disc where the stream was stuck in a 12+ hour bad-sector grind.

0.11.17 (2026-04-23)

Adaptive batch sizer in DiscStream — no more per-sector descent

Rip recovery rewritten. The old binary-search-per-bad-sector model paid the full descent (batch → half → quarter → … → single) for every bad sector in a region. On a damaged disc with 600 consecutive bad sectors this took 12+ hours. The new algorithm pays the descent once, remembers the working size, and ramps back up only after a sustained clean streak.

• BatchSizeChanged { new_size, reason } event — fires on shrink (read failed) and probe-up (clean streak threshold hit). Consumers use this to distinguish a "recovering" rip from a normal one.
• Removed BinarySearch and SectorRecovered emissions from DiscStream — no longer produced by the rip path. SectorRecovered still fires from Drive::read's multi-phase recovery (unused by rips today, but kept for scan/other callers).
• Removed read_with_binary_search and the 3×5s light-recovery loop — no retry loops, no sleeps. One 5s attempt per read. On size-1 failure, skip (zero-fill) or error.
• Probe-up threshold: 100 MiB (51,200 sectors) of clean reading at current size before doubling toward preferred. Ramp 1 → preferred on good reading takes ~100 seconds for a typical BD — trivial vs. rip duration, conservative enough that a single lucky sector in a marginal zone can't trigger a premature probe.
• Bad-region math: ~600 consecutive bad sectors now complete in ~50 min (600 × 5s) instead of ~12h. The descent is O(log preferred) one time, not per sector.

macOS

• Fix new clippy lint (manual_c_str_literals) in scsi/macos.rs.

0.11.16 (2026-04-21)

API cleanup — one method per action

• SectorReader::read_sectors(lba, count, buf, recovery) — single method with recovery: bool. Removes read_sectors_recover().
• parser_for_codec(codec, codec_data) — single constructor. Removes parser_for_codec_with_data().
• DvdSubParser::new(codec_data) — single constructor. Removes with_codec_data().
• MkvMuxer::new(writer, tracks, title, duration, chapters) — single constructor. Removes new_with_chapters().

0.11.15 (2026-04-21)

Lint cleanup

• Fix all cargo fmt and cargo clippy -D warnings across codebase.
• Remove unused imports, dead code, collapsible if-statements, div_ceil reimplementation.

0.11.14 (2026-04-21)

Audit fixes: read recovery, verify, SCSI

• Fix: trailing sectors at extent boundaries — extents with sector_count not divisible by 3 no longer drop 1-2 trailing sectors. decrypt_sectors() safely skips partial AACS units.
• Fix: verify_title stop support — progress callback now returns bool. Return false to stop verification early instead of running to completion.
• Fix: O_CLOEXEC on all SCSI fd opens — prevents fd leak to child processes.
• Fix: SCSI sense descriptor format — correctly detect response code 0x72/0x73 (descriptor format) and extract sense key from byte 1 instead of byte 2.
• Fix: DecryptFailed on missing unit key — decrypt_sectors() returns Err(DecryptFailed) instead of silently using a zero key.

0.11.13 (2026-04-21)

Fix: all rip reads use fast timeout

• Initial batch read changed from full Drive::read() recovery to fast 5s timeout. Binary search starts immediately on failure instead of after 10 minutes of retries.
• Max 15 seconds per bad sector (3 x 5s attempts). Max 23 seconds per batch with 1 bad sector.

0.11.12 (2026-04-21)

Drive halt + sector events + light recovery

• Drive.halt() — AtomicBool flag checked between retry phases. Max 30s to stop.
• Drive.on_event() — callback for ReadError, Retry, SpeedChange, SectorRecovered events.
• Error::Halted (E6010) — distinct from DiscRead, indicates intentional stop.
• Binary search light recovery — single sectors get 3 attempts x 5s (15s max) instead of full 10-min Drive::read() recovery. Marginal disc zones complete in minutes not hours.
• DiscStream.on_event() — BinarySearch, SectorRecovered, SectorSkipped events.

0.11.11 (2026-04-20)

Binary search error recovery

• fill_extents binary search — when a batch read fails, binary search to isolate the failing sector(s). Good sectors read in sub-batches at full speed. Only truly bad sectors get individual recovery. 60-sector batch with 1 bad sector: ~5 seconds instead of 10+ minutes.

0.11.10 (2026-04-20)

Skip errors + clean verify API

• DiscStream.skip_errors — when true, zero-fills unreadable sectors and continues instead of aborting. Caller sets based on user preference.
• read_sectors_recover(recovery: bool) — single API for recovery vs fast reads. Replaces separate read_sectors_fast method.

0.11.9 (2026-04-20)

Fast verify reads

• read_sectors_fast() — single-attempt 5s timeout SCSI read for verify. No recovery loop. Bad sectors detected in seconds instead of 10+ minutes.
• SectorReader trait — added read_sectors_fast() with default fallback to read_sectors().

0.11.8 (2026-04-20)

Disc verify

• verify::verify_title() — sector-by-sector health check. Classifies sectors as Good/Slow/Recovered/Bad. Progress callback, chapter mapping, sector ranges.

0.11.7 (2026-04-19)

TrueHD parser rewrite

• 12-bit length mask — access unit length is lower 12 bits of first 2 bytes, not full 16. Upper 4 bits are parity nibble. Wrong mask caused misaligned frame splits.
• AC-3 frame skipping — BD-TS TrueHD PES contains interleaved AC-3 frames (same PID). Parser now detects AC-3 sync word (0x0B77) and skips those frames.
• Cross-PES buffering — access units that span PES packet boundaries are correctly reassembled.
• Per-unit timestamps — each access unit gets incrementing PTS (1/1200th second apart) instead of all units in one PES sharing the same timestamp.
• Major sync detection — keyframe flag set when access unit contains MLP major sync (0xF8726FBA).
• Result: zero TrueHD decode errors on UHD and BD (was ~19 per 30 seconds).

0.11.6 (2026-04-18)

TrueHD fix (incomplete)

• Initial attempt at TrueHD header stripping — wrong approach, superseded by 0.11.7.

0.11.5 (2026-04-18)

MKV container fixes — Jellyfin/player compatibility

• Timestamp normalization — MKV and M2TS output starts at 0.000s instead of raw disc PTS offset. Fixes playback failures in Jellyfin and other players.
• DefaultDuration — correct frame rate written to MKV track header. Fixes wrong avg_frame_rate (was 293/12, now 24000/1001).
• HDR Colour metadata — MatrixCoefficients, TransferCharacteristics, Primaries, Range written to MKV video track. Enables HDR tone mapping in players.
• DisplayWidth/DisplayHeight — aspect ratio fields in MKV video track.
• Chapters (Blu-ray) — accept mark_type 0 as chapter entry (was filtering to type 1 only, which no disc uses).
• Chapters (DVD) — extract chapter timestamps from PGC program map + cell durations.
• Default disposition — only first video and first audio track marked default. Fixes wrong auto-selection in players.

0.11.3 (2026-04-18)

Unified versioning

• All freemkv repos now share the same version number. No functional changes from 0.10.10.

0.10.10 (2026-04-18)

Dual-layer disc fix

• UDF extent allocation — use actual UDF allocation descriptors (file_extents()) instead of assuming m2ts files are contiguous from file_start_lba. Dual-layer UHD discs split large files across many extents (~1 GB each). The old single-extent assumption truncated rips at ~37% on affected discs.
• Read error propagation — fill_extents() returns io::Result<bool> so SCSI read errors propagate to the caller instead of being silently treated as EOF.

0.10.9 (2026-04-17)

Fast disc identification

• Disc::identify() — reads UDF filesystem only (name, format, layers, encrypted). ~3s on USB vs 18s for full scan. No AACS handshake or playlist parsing.
• KEYDB path fix — added ~/.config/freemkv/keydb.cfg to search paths. Fixes silent rip hang when KEYDB exists but isn't found by resolve_keydb().

0.10.8 (2026-04-17)

Buffered UDF reads

• BufferedSectorReader — prefetches batch sectors on single-sector reads. USB drives have ~500ms per SCSI command; this eliminates scan hangs.
• Metadata partition pre-read — loads entire UDF metadata partition into memory after initial parse.
• Scan time reduced from 10+ minutes to ~18 seconds on USB.

0.10.7 (2026-04-17)

DiscStream::new()

• Replaced open_drive(), open_iso(), from_reader() with single new() constructor
• Stream accepts ContentFormat and sets up demuxer internally
• Removed disc:// case from input() — callers use primitives directly

0.10.6 (2026-04-16)

Docker compatibility

• Drive discovery — removed sysfs check that blocked detection inside Docker containers. Device nodes are sufficient; INQUIRY command validates the device is an optical drive.

0.10.5 (2026-04-16)

Audio parser buffering

• AC3 — buffer across PES boundaries with frame size from fscod/frmsizecod table. Eliminates all AC3 decode errors on BD and UHD.
• DTS — buffer with core sync detection + frame size from header. DTS-HD extension frames handled correctly.
• TrueHD — buffer with unit length field parsing. Incomplete units held for next PES.
• All audio parsers now emit complete frames only. When PES boundaries align (normal case), buffering is a no-op.

0.10.4 (2026-04-16)

CSS decryption — full key hierarchy

• Bus auth → disc key → title key — complete CSS key chain. Bus authentication with CSSCryptKey challenge-response, disc key decryption using 31 player keys via READ DVD STRUCTURE, title key extraction via REPORT KEY format 0x04.
• CSS descramble cipher — correct LFSR keystream generation with TAB5 for LFSR1 output and TAB4 for LFSR0 output. Per-sector key derivation from title key XOR sector seed.
• Stevenson plaintext attack — expanded pattern set (padding, video, audio, nav pack headers), scans up to 50K scrambled sectors for ISO key recovery.
• Disc::copy() CSS decrypt — sector-level decryption during disc→ISO copy produces clean ISOs with zero scramble flags.

MPEG-2 PS demuxer fixes

• DVD PS path routes through codec parsers — was bypassing parser.parse(), producing raw PES frames without codec_private extraction or keyframe detection.
• MPEG-2 sequence header extraction — calculates exact header size including quantizer matrices (intra/non-intra flags), captures sequence extension from subsequent PES packets.
• TsDemuxer dynamic PID table — Vec instead of fixed [i16; 8192] for DVD PIDs that may exceed 8192.

0.10.3 (2026-04-16)

DVD CSS authentication

• CSS drive authentication — full SCSI REPORT KEY / SEND KEY handshake with 6-round substitution-permutation cipher (CSSCryptKey). Brute-forces variant from 32 possibilities. Drive serves scrambled sectors after auth completes.
• CSS auth runs before scan — chicken-and-egg fix: auth must happen before reading VOB sectors for title key cracking, not after.
• Remove debug output — strip temporary eprintln from drive reads and CSS auth.

0.10.2 (2026-04-15)

Fixes

• Disc::copy() batch overflow — hardcoded 64-sector batch exceeded BU40N's 60-sector hardware limit, causing every read to fail and trigger 5×30s recovery sleep. Now accepts detected batch size from caller, defaults to 60.
• IFO PGC parsing — playback time read from offset 0x04 (correct) instead of 0x02 (nr_programs). Cell BCD time at cell+4 not cell+0. DVD durations now correct.
• Demuxer flush at EOF — TS and PS demuxers flushed when source reaches EOF, preventing loss of last PES frame. Applied to DiscStream and M2tsStream.
• DiscStream demuxer selection — demuxer set by caller based on content_format (TS for Blu-ray, PS for DVD) instead of unconditionally creating TsDemuxer in from_reader()
• StdioStream FMKV header — writes/reads metadata header for roundtrip compatibility through stdio pipes

0.10.1 (2026-04-15)

Architecture: streams are PES, disc.copy() for sector dumps

• One stream per format, bidirectional PES — MkvStream, M2tsStream, NetworkStream, StdioStream, NullStream each handle read and write
• IsoStream merged into DiscStream — one type for physical drives and ISO files, different SectorReader
• Disc::copy() — raw sector dump for disc→ISO, not a stream operation
• IOStream deleted — no more byte-level Read/Write on streams
• ContentReader/OpenDisc deleted — replaced by DiscStream + PES pipeline
• CountingStream — wrapper for progress tracking, no state in streams

Error codes only — zero English in library

• All io::Error::new(kind, "english") replaced with Error enum variants
• New error variants: StreamReadOnly, StreamWriteOnly, StreamUrlInvalid, MkvInvalid, NoStreams, etc.
• From<Error> for io::Error — clean conversion at system boundaries
• Removed unused error variants: WriteError, ProfileNotFound, NotUnlocked, NotCalibrated, ScsiTimeout, etc.

Deleted dead code

• mkvout.rs, pesout.rs, isowriter.rs — merged into parent stream types
• lookahead.rs usage in MkvStream — replaced by PES direct write
• ContentReader, OpenDisc, open_title() — replaced by PES pipeline
• open_input(), open_output() — replaced by input(), output()

0.10.0 (2026-04-15)

PES pipeline

• Unified Stream trait — read() returns PES frames, write() accepts them. One trait for all streams.
• All streams produce/consume PES frames — DiscStream, IsoStream, MkvStream, M2tsStream, NetworkStream, StdioStream, NullStream
• DVD PS demux — MPEG-2 Program Stream demuxer produces PES frames
• MKV input stream — MKV demux produces PES frames
• Network/stdio PES — PES serialization over TCP and pipes
• FileSectorReader — ISO files implement SectorReader for unified disc/ISO handling

PES pipeline audit (20 fixes)

• PES serialize: track/length validation, OOM cap (256 MB), stuffing compliance
• TsDemuxer: AF length validation, find_start_code verified
• PTS: marker bit validation, ns→90kHz saturating_mul, round-to-nearest
• AC3/DTS: debug_assert promoted to runtime check
• MKV: block_vint 3-4 byte support, track bounds check
• FMKV: JSON 10 MB cap, PAT section_len underflow guard

codec_privates refactor

• codec_privates on DiscTitle — no separate parameter passing, no _with_X method variants
• Streams-not-files — MkvStream and M2tsStream take impl Read, not File/Seek
• M2TS roundtrip fix — TsMuxer Annex B conversion + codec_private in FMKV header
• MKV remux fix — MkvStream returns codec_privates from EBML header
• Network codec_private fix — FMKV header carries base64 codec_privates

Cleanup

• Remove Seek/File dependencies from stream interfaces
• Remove eprintln from library code
• Fix all clippy warnings
• 342 tests pass

0.9.0 (2026-04-14)

Drive recovery + decrypt architecture

• Drive::read() — single read method with built-in error recovery (min speed → reset → retry)
• Decrypt in streams — streams handle their own decryption via decrypt_sectors(). Pipeline just moves bytes.
• keys() on IOStream — streams report their own decrypt keys
• InputOptions — --raw wired through to streams, skips decrypt only
• decrypt_sectors returns Result — fail instead of silent corruption
• Handshake fix — no longer returns fake success on failure
• Drive::read_capacity() — for raw sector dump (disc→ISO)
• Reset on open — SgIoTransport resets device on every open
• Simplified DiscStream — removed on_error/on_success/Recovery enum

Platform

• Rust 1.86 MSRV pinned in Cargo.toml and CI
• macOS build fix — MacScsiTransport marked Send
• is_multiple_of — replaced nightly API with stable equivalent

API changes

• Drive object — typed DriveSession API
• Typed StreamUrl — URL parsing returns enum, not strings
• DriveStatus API — reset(), wait_ready with fallback
• Granular SCSI queries — individual methods on DriveSession for capture
• Profile module public — for external tools (bdemu)
• Tray lock/unlock — exposed on Drive

0.8.0 (2026-04-11)

DVD support

• Full DVD pipeline — VIDEO_TS detection, IFO parsing, CSS decryption, MPEG-2 PS demuxing
• CSS cipher — Stevenson 1999 table-driven implementation, no keys needed
• IFO parser — title sets, PGC chains, cell addresses, audio/subtitle attributes, palette
• MPEG-2 PS demuxer — pack headers, PES extraction, private stream 1 sub-streams
• MPEG-2 video parser — sequence headers, I-frame detection, codec_private

100% codec coverage

• E-AC-3 (Dolby Digital Plus) — bsid detection, frame size calculation
• DTS-HD MA/HR — extension substream detection and inclusion
• LPCM — BD header skip, raw PCM extraction
• DVD subtitles (VobSub) — passthrough with IFO palette extraction (YCbCr→RGB)
• Dolby Vision — verified RPU NAL type 62 preserved in HEVC passthrough

MKV improvements

• Chapters — MPLS PlayList marks → MKV Chapters element
• Track flags — FlagDefault, FlagForced, Language correctly set
• HEVC codec_private — profile compatibility and constraint flags from SPS
• VC-1 codec_private — resolution parsed from sequence header

Architecture

• SectorReader trait — decouples disc scanning from SCSI
• Disc::scan_image() — scan ISO images or any SectorReader
• resolve_encryption() — single function handles AACS 1.0/2.0/CSS/none
• Module refactors — disc/ (4 files), aacs/ (5 files), drive/ (3 files)
• Module visibility — internal modules pub(crate), explicit AACS re-exports

Streams

• StdioStream — stdin/stdout pipe
• IsoStream — read/write Blu-ray ISO images with UDF 2.50 filesystem
• Strict URLs — all URLs require scheme:// prefix, bare paths rejected
• total_bytes() — IOStream reports content size for progress display

Platform

• Windows SPTI — SCSI Pass-Through Interface backend
• Windows builds — CI + release workflow for x86_64-pc-windows-msvc
• macOS drive discovery — separate from Linux (drive/macos.rs)
• Stable download URLs — /latest/download/ with version-free filenames

Audit fixes (4 rounds, 14→0 critical)

• UDF bounds checking on all disc-sourced offsets
• SCSI: Linux residual underflow, macOS task_status type, Windows buffer zeroing
• AACS: EC mod_inv safe, key reduced mod n, host cert fallback
• DiscStream: persistent read state (was recreating ContentReader per call)
• ISO writer: UDF tag checksums, multi-extent >4GB, reserve AVDP placement
• CSS crack: labeled loop break, polynomial match
• 0 clippy warnings

Testing

• 327 tests (was 64 at start)
• CSS/AACS cross-validation against independent AES implementation
• End-to-end MKV mux test with H.264 codec headers

0.7.2 (2026-04-11)

Windows support

• SPTI backend (scsi/windows.rs) — SCSI_PASS_THROUGH_DIRECT via DeviceIoControl
• Windows drive discovery (drive/windows.rs) — scans CdRom0-15 + drive letters
• Platform file separation — drive/unix.rs and drive/windows.rs, no inline cfg branches
• CI — cargo check on windows-latest, actions/checkout@v5

Test suite

• 177 tests (was 64) — MPLS, CLPI, H.264, HEVC, AC3, VC1, DTS, TrueHd, PGS, EBML, UDF, disc scanning, streams
• FEATURES.md created

Improvements

• Stable download URLs — /latest/download/freemkv-x86_64-unknown-linux-musl.tar.gz works forever

0.7.1 (2026-04-11)

SectorReader trait

• SectorReader trait — decouples disc scanning from SCSI. UDF, MPLS, CLPI, labels, and AACS resolution now work with any sector source.
• Disc::scan_image() — scan ISO images or any SectorReader. Full title/stream/label/AACS pipeline, no drive required.
• resolve_encryption() — single function handles AACS 1.0, 2.0, or none. Uses whatever path works (KEYDB VUK, handshake, media key, device key).

Stream types

• 7 stream types — Disc, ISO, MKV, M2TS, Network, Stdio, Null
• IsoStream — read/write Blu-ray ISO images. Uses Disc::scan_image() for full UDF parsing (not heuristic scanning).
• StdioStream — stdin/stdout pipe, format-agnostic
• Strict URL format — all URLs require scheme://path. Bare paths rejected with clear error messages.
• Validation — empty paths, missing ports, read-only/write-only direction errors

IOStream trait

• IOStream trait for all stream types (Read + Write + info + finish)
• open_input() / open_output() resolve URL strings to stream instances

0.7.0 (2026-04-11)

Stream I/O architecture

• 5 stream types — Disc, MKV, M2TS, Network, Null
• IOStream trait — common interface for all streams
• URL resolver — open_input() / open_output() with scheme://path format
• FMKV metadata header — JSON metadata embedded in M2TS and network streams
• Bidirectional MKV — MkvStream reads and writes Matroska containers
• Network streaming — TCP with metadata header, TCP_NODELAY
• BD-TS demuxer — PAT/PMT scanning, PTS duration detection
• EBML reader — parse existing MKV files for read-side MkvStream

0.6.0 (2026-04-10)

API improvements

• open() works on all drives — no profile match required. Unknown drives can scan, read BD/DVD at OEM speed. init() is optional and adds features (riplock removal, UHD reads, speed control).
• has_profile() — check if unlock parameters are available for this drive
• find_drives() — returns all optical drives, not just profile-matched ones
• raw_gc_010c on DriveId — raw GET_CONFIG 010C response bytes for profile sharing

AACS 2.0

• SCSI handshake wired end-to-end — ECDH key agreement, real Volume ID from drive, read data key for bus decryption
• Bus decryption active — UHD discs with bus encryption now decrypted transparently
• VUK derivation from Media Key + VID — works for discs not in KEYDB (processing key + device key paths)

MKV muxer

• 15 new files — EBML writer, TS demuxer, stream assembly pipeline
• Codec parsers — H.264, HEVC, AC-3, DTS, TrueHD, PGS, VC-1
• MkvStream — builder pattern, wraps any impl Write, configurable lookahead buffer

Cleanup

• Removed orphaned jar.rs (342 lines) — replaced by labels/ module
• Error refactor: 40+ sites converted from English strings to typed error codes

0.5.0 (2026-04-09)

Read pipeline — 5x speed improvement

• Kernel transfer limit detection: auto-detect max_hw_sectors_kb via sysfs, resolve sg→block device. Previously hardcoded to 510 sectors (1MB) which exceeded the 120KB kernel limit, causing all reads to error and fall back to 6KB reads at 4.8 MB/s. Now auto-tunes to 48 sectors (96KB) or whatever the device supports.
• Result: 12.5 MB/s sustained, 23 MB/s peak (was 4.8 MB/s)

Drive init pipeline

• Full drive init pipeline: unlock, calibrate (256 zones), register reads, status, probe, set_read_speed, keepalive, timing
• Speed calibration: 256+ disc surface probes, 64-entry speed table, triple SET_CD_SPEED
• Platform trait locked down: pub(crate), 3 methods only (init, set_read_speed, is_ready)
• Init guard: prevents double-init, signature mismatch aborts early

MPLS parser fixes

• PGS in audio slots: subtitle language read at correct offset (was truncated: "ng " → "eng")
• Secondary PG entries: n_pip_pg loop added for correct STN position tracking
• Secondary stream types: stream_type 5 (sec audio), 6 (sec video), 7 (DV EL) attribute parsing
• Empty stream filter: coding_type 0x00 entries (padding) no longer appear as "Unknown(0)"

Profiles

• 206 profiles with full per-drive data: ld_microcode (base64), all CDBs, speed tables, signatures
• Automated pipeline: profile import → profiles.json (no manual merging)

0.4.0 (2026-04-07)

Labels — complete rewrite

• Detect-then-parse architecture: each BD-J authoring format has its own parser module with detect() and parse() functions. Drop in a new parser with one line in the registry.
• 5 format parsers: Paramount (playlists.xml), Criterion (streamproperties.xml), Pixelogic (bluray_project.bin), Warner CTRM (menu_base.prop / language_streams.txt), shared label vocabulary (vocab.rs)
• Raw disc data principle: label data passes through as-is from disc. Only BD-standard codec identifiers (MLP, AC3, DTS) are mapped to display names. Unknown authoring tool codes (csp, eda, cf) pass through raw.
• variant field: replaces region — language dialect codes from authoring tools, not BD spec regions
• Removed: old jar module (superseded by labels)
• Removed: dead label apply functions from disc.rs

Drive

• DriveSession::eject(): sends PREVENT ALLOW MEDIUM REMOVAL then START STOP UNIT. Works reliably after raw mode unlock.
• DiscRegion enum: Free, BluRay(A/B/C), Dvd(1-8). UHD always region-free.

Capture

• Fixed sector range collection: captures ALL files on disc (only skips STREAM/ video files and >50MB). Previously skipped BACKUP/, DUPLICATE/, and files >10MB which missed JAR content.

0.3.1

• Labels module: 4 disc file parsers for stream labels
• Simplified labels API

0.3.0

• Initial public release
• SCSI transport (Linux SG_IO, macOS IOKit)
• UDF 2.50 filesystem reader
• MPLS/CLPI parsers with full STN support
• Drive identification + profile matching
• 206 bundled drive profiles
• AACS 1.0 decryption (VUK + unit keys)