diff --git a/quest/m0/ci/README.md b/quest/m0/ci/README.md index 91a5e25babd..542babedf06 100644 --- a/quest/m0/ci/README.md +++ b/quest/m0/ci/README.md @@ -163,4 +163,6 @@ The historical samples locate candidates; they are not controlled benchmarks. - [Fewer BTreeMap instantiations](/quest/m0/ci/btree-keys.md) - moq-net shares BTreeMap key and value types, kept only if its CPU time drops - [Halve the slowest tests' CPU](/quest/m0/ci/slow-tests.md) - the slowest 1% of tests take under half their 289 CPU-seconds, with coverage kept - [Release credentials off main](/quest/m0/ci/publish-release-only.md) - `publish` admits only `release` and release tags once its `workflow_run` publishers run on their tag +- [The queue host keeps the queue's disk](/quest/m0/ci/queue-host-disk.md) - agent sessions can't starve the queue runner of disk, and a crashed runner restarts +- [The next test-CPU cuts](/quest/m0/ci/test-cost-followups.md) - the slow-test leftovers from #6987, cut at their source - [Biome checks every file it can parse](/quest/m0/ci/biome-whole-repo.md) - Biome runs over the whole repository and main is clean under it diff --git a/quest/m0/ci/queue-host-disk.md b/quest/m0/ci/queue-host-disk.md new file mode 100644 index 00000000000..99436bf2f05 --- /dev/null +++ b/quest/m0/ci/queue-host-disk.md @@ -0,0 +1,32 @@ +# [M] The queue host keeps the queue's disk + +## Goal + +Agent sessions sharing the self-hosted queue host can never fill its disk to +the point where the `moq-queue` runner fails a check, and a runner that does +crash comes back on its own. + +## Plan + +On 2026-10-30 the host filled twice (about 03:00 and 07:30 PDT). Each time +non-CI usage grew by roughly 480 GB and was later freed. CI's XFS quota is +whatever the host leaves free past a 32 GiB reserve, so the runner hit ENOSPC +mid-check: quest land ejected moq-dev/moq#6985 and moq-dev/moq#6999 as +`failed_checks`, and at 07:49 the runner aborted and stayed down for 36 +minutes. + +Done so far, in the host's configuration: +an `agent-storage-gc` user timer reclaims idle worktree build outputs and +leaked test scratch every 10 minutes and logs free space, and the runner unit +restarts after a crash (that change is in the PR but not yet deployed). + +Remaining: + +- Find what consumed the 480 GB, using the timer's free-space log the next + time it happens, and fix that source. +- Guarantee CI a floor: put agent storage (`/home`, `/tmp`) in its own XFS + project with a hard cap, so an agent gets ENOSPC instead of the queue. + +Decided 2026-10-30 (maintainer, /quest-plan of the 2026-10-30 /quest-complete follow-ups): ✅ new m0 quest / note in the host configuration only / drop. + +Public API: none. Wire: none. diff --git a/quest/m0/ci/test-cost-followups.md b/quest/m0/ci/test-cost-followups.md new file mode 100644 index 00000000000..74844081cc3 --- /dev/null +++ b/quest/m0/ci/test-cost-followups.md @@ -0,0 +1,25 @@ +# [M] The next test-CPU cuts + +## Goal + +The test costs [moq-dev/moq#6987](https://github.com/moq-dev/moq/pull/6987) +found but left are cut at their source, with coverage kept. + +## Plan + +Left by #6987 (halve the slowest tests' CPU): + +- Encode the recovery clips once per run instead of per test. +- Use `tracing-test`'s `no-env-filter` where its filter costs time. +- Make the moq-stats oversize tests cheap. +- Make `a_fill_refused_group_by_group_stays_linear` cheap without losing what + it proves. + +Measure each before and after, as #6987 did. The export idle poll and mesh +session cost already have quests in `quest/m2/perf/`, and +[moq-dev/moq#7075](https://github.com/moq-dev/moq/pull/7075) already decodes a +moq-quic padding run as one frame. + +Decided 2026-10-30 (maintainer, /quest-plan of the 2026-10-30 /quest-complete follow-ups): ✅ new m0 CI quest / drop. + +Public API: none. Wire: none. diff --git a/quest/m0/test-flakes-2/README.md b/quest/m0/test-flakes-2/README.md index 5b58ad6b455..f2b20490eed 100644 --- a/quest/m0/test-flakes-2/README.md +++ b/quest/m0/test-flakes-2/README.md @@ -57,3 +57,5 @@ Public API: none. Wire: none. ## Required - [IETF22 fallback](/quest/m0/test-flakes-2/ietf22-fallback.md) - an IETF22 browser viewer follows a fallback within the media lane's bound, keeping its session and its audio +- [moq-relay's auth-trust test survives load](/quest/m0/test-flakes-2/relay-auth-trust.md) - it stops timing out on a busy host +- [Draft-22 interop with a native publisher](/quest/m0/test-flakes-2/draft22-interop.md) - no "invalid fetch range" or audio gap with a Rust or C++ publisher diff --git a/quest/m0/test-flakes-2/draft22-interop.md b/quest/m0/test-flakes-2/draft22-interop.md new file mode 100644 index 00000000000..5d826b95ee2 --- /dev/null +++ b/quest/m0/test-flakes-2/draft22-interop.md @@ -0,0 +1,20 @@ +# [S] Draft-22 interop with a native publisher holds + +## Goal + +Interop on moq-transport draft 22 with a Rust or C++ publisher stops failing +with "invalid fetch range" and an audio gap, or is shown not to reproduce on +current main. + +## Plan + +[moq-dev/moq#7085](https://github.com/moq-dev/moq/pull/7085) hit it and cited +`ffi-audio-gap.md`, which [moq-dev/moq#6990](https://github.com/moq-dev/moq/pull/6990) +deleted. [moq-dev/moq#7204](https://github.com/moq-dev/moq/pull/7204) has +since limited IETF publishers to one group of history, which may change it. +Reproduce on current main first. If it no longer reproduces, delete this quest +with that evidence; otherwise fix the range at its source. + +Decided 2026-10-30 (maintainer, /quest-plan of the 2026-10-30 /quest-complete follow-ups): ✅ new flake child / drop. + +Public API: none. Wire: none unless the fix shows a FETCH range bug. diff --git a/quest/m0/test-flakes-2/relay-auth-trust.md b/quest/m0/test-flakes-2/relay-auth-trust.md new file mode 100644 index 00000000000..94dadcb7ff8 --- /dev/null +++ b/quest/m0/test-flakes-2/relay-auth-trust.md @@ -0,0 +1,17 @@ +# [XS] moq-relay's auth-trust test survives load + +## Goal + +`http_client::tests::auth_trust_is_independent_from_mesh_verification` in +moq-relay stops timing out when the host is busy, fixed at its cause. + +## Plan + +Seen during [moq-dev/moq#6994](https://github.com/moq-dev/moq/pull/6994)'s +local `just ci`: it timed out under load and passes alone in 0.08 s. Find what +it waits on by wall clock and make it deterministic, rather than raising a +timeout. + +Decided 2026-10-30 (maintainer, /quest-plan of the 2026-10-30 /quest-complete follow-ups): ✅ new flake child / drop. + +Public API: none. Wire: none. diff --git a/quest/m1/README.md b/quest/m1/README.md index 10c46b93117..38b62fe1570 100644 --- a/quest/m1/README.md +++ b/quest/m1/README.md @@ -171,6 +171,7 @@ removed). - [Next Object on a cold Rust copy](/quest/m1/next-object-cold-copy.md) - a join after an upstream group report starts at the first frame received, never skipping - [Legacy lite close drain](/quest/m1/lite-legacy-close-drain.md) - a lite-01..06 graceful close never cuts a final group the subscriber has not read - [moqsrc max-delay](/quest/m1/gst-max-delay.md) - GStreamer's `moqsrc` takes a `max-delay` property instead of a hard-coded 1 s budget +- [JS join tests cover draft 20](/quest/m1/join-tests-draft20.md) - join.test.ts runs on every supported draft, not just 18 and 22 - [Poll-first async](/quest/m1/poll-first/README.md) - library crates compile no coroutines by default: async wrappers are generated from `poll_*` and gated behind an `async` feature - [Binding announcements always carry captures](/quest/m1/ffi-required-captures.md) - moq-ffi and its generated bindings make `captures` a required list, as #7266 always sets it - [HLS export follows a resolution change](/quest/m1/hls-reinit-same-codec.md) - a decoder-identity change under the same codec string starts a new init and segment tag diff --git a/quest/m1/api-audit/README.md b/quest/m1/api-audit/README.md index e09056d35b3..0b87b2e9ed7 100644 --- a/quest/m1/api-audit/README.md +++ b/quest/m1/api-audit/README.md @@ -77,6 +77,7 @@ Rejected in planning: - [Routes are the origin API](/quest/m1/api-audit/route/README.md) - apps announce exact or prefix routes through one `Route` handle that owns its broadcast and carries an `identity`, prefix routes can serve a held broadcast, and readers get a listing plus `follow` - [No constructible internal handles](/quest/m1/api-audit/internal-constructors.md) - `Rendition`, `Lister`, `Arrivals`, and e2ee's generations get private constructors, and a lint rule stops new ones - [moq-congestion's first-publish names](/quest/m1/api-audit/congestion-names.md) - `bbr3::Config`, `cubic::Config`, `Metrics`, `Factory`, and no `into_any` +- [@moq/net names the origin read side](/quest/m1/api-audit/origin-reader.md) - `Origin.Reader`, which `Origin.Table` extends, replaces `Pick` lists ## Related diff --git a/quest/m1/api-audit/origin-reader.md b/quest/m1/api-audit/origin-reader.md new file mode 100644 index 00000000000..bab430313fd --- /dev/null +++ b/quest/m1/api-audit/origin-reader.md @@ -0,0 +1,20 @@ +# [S] @moq/net names the origin read side + +## Goal + +@moq/net exports a named read-side origin interface, `Origin.Reader`, that +`Origin.Table` extends, so readers such as @moq/watch accept that one name +instead of repeating a `Pick` list. + +## Plan + +[moq-dev/moq#7251](https://github.com/moq-dev/moq/pull/7251) makes @moq/watch +accept a structural read-side origin and leaves the named interface as a +follow-up. Name it before the API cut, and move existing `Pick` lists to it. +If #7251 has not merged, `Player` and `Broadcast` still take `Origin.Table`: +widen their origin inputs to `Origin.Reader` here. + +Decided 2026-10-30 (maintainer, /quest-plan of the 2026-10-30 /quest-complete follow-ups): ✅ new m1 api-audit quest / drop. + +Public API: @moq/net adds `Origin.Reader`; @moq/watch's origin inputs accept +it. Both are additive, so nothing migrates. Wire: none. diff --git a/quest/m1/c/api-gaps.md b/quest/m1/c/api-gaps.md index e545717de76..36433e189a5 100644 --- a/quest/m1/c/api-gaps.md +++ b/quest/m1/c/api-gaps.md @@ -3,15 +3,16 @@ ## Goal moq-c reports a malformed epoch the same way everywhere, lets a C consumer see -that audio is disabled as it can for video, and has a test of the hops on a -route received through a relay. +that audio is disabled as it can for video, has a test of the hops on a route +received through a relay, stamps a JSON window entry with a caller's timestamp, +and can disable a simulcast rung on pre-encoded media. ## Plan Found 2026-10-28 by [moq-dev/moq#7151](https://github.com/moq-dev/moq/pull/7151) and [moq-dev/moq#7156](https://github.com/moq-dev/moq/pull/7156). -Decided 2026-10-28 (maintainer): one PR for all three. moq-c already breaks +Decided 2026-10-28 (maintainer): one PR for all of these. moq-c already breaks at this cut (0.6 to 0.7), so the error-code change rides that break. - `moq_broadcast_request_prepare` reports a malformed epoch as @@ -25,8 +26,21 @@ at this cut (0.6 to 0.7), so the error-code change rides that break. - Add a moq-c test that announces through a relay session and checks the received route's `hops`. +- `moq_publish_json_window_push` and `_pop` always stamp the broadcast clock. + Let them take an optional timestamp like the JSON snapshot and stream + writers do since [moq-dev/moq#6946](https://github.com/moq-dev/moq/pull/6946). + Decided 2026-10-30 (maintainer, /quest-plan of the 2026-10-30 /quest-complete follow-ups): + ✅ add it here / drop. +- Add C ABI support for disabling simulcast rungs on pre-encoded media, which + [moq-dev/moq#7029](https://github.com/moq-dev/moq/pull/7029) added to moq-ffi + for OBS. Decided 2026-10-30 (maintainer, /quest-plan of the 2026-10-30 /quest-complete follow-ups): the C epic's rule (every m1 + binding feature missing from `moq.h` lands before the cut) overrides + #7029's decision 4, which parked it until a C consumer asked: + ✅ add to m1 C gaps / exempt it. + Update `moq.h`, `rs/moq-c/README.md` and `doc/lib/c` as needed. -Public API: moq-c gains an audio enabled query and changes one error code. -Wire: none. +Public API: moq-c gains an audio enabled query, an optional caller timestamp +on JSON window push and pop, and a call to disable a rung of pre-encoded +media, and changes one error code. Wire: none. diff --git a/quest/m1/join-tests-draft20.md b/quest/m1/join-tests-draft20.md new file mode 100644 index 00000000000..e5ae1a7f324 --- /dev/null +++ b/quest/m1/join-tests-draft20.md @@ -0,0 +1,19 @@ +# [XS] JS join tests cover draft 20 + +## Goal + +`js/net/src/ietf/join.test.ts` runs its join cases on every moq-transport +draft @moq/net supports, including draft 20, instead of hard-coding drafts 18 +and 22. + +## Plan + +[moq-dev/moq#7190](https://github.com/moq-dev/moq/pull/7190) restored draft 20 +and added it to `fetch.test.ts`, but `join.test.ts` still names drafts 18 and +22. Iterate its cases over the supported versions, as `fetch.test.ts` does, +with each draft's own expectation: draft 18 joins with a relative joining +FETCH and draft 22 with a fill, so confirm which path draft 20 takes first. + +Decided 2026-10-30 (maintainer, /quest-plan of the 2026-10-30 /quest-complete follow-ups): ✅ new m1 XS quest / drop. + +Public API: none. Wire: none. diff --git a/quest/m1/push-to-talk/README.md b/quest/m1/push-to-talk/README.md index 7d30c45da28..e32a03eda08 100644 --- a/quest/m1/push-to-talk/README.md +++ b/quest/m1/push-to-talk/README.md @@ -55,6 +55,7 @@ encoder, and record why. ## Required - [Watch caches audio codec support](/quest/m1/push-to-talk/audio-support-cache.md) - an unmute resumes audio without waiting on a support check +- [Hidden capture idles](/quest/m1/push-to-talk/capture-idle.md) - an unread capture stops producing frames but stays open for an instant unhide ## Closes diff --git a/quest/m1/push-to-talk/capture-idle.md b/quest/m1/push-to-talk/capture-idle.md new file mode 100644 index 00000000000..97003e3ffc0 --- /dev/null +++ b/quest/m1/push-to-talk/capture-idle.md @@ -0,0 +1,19 @@ +# [S] Hidden capture idles + +## Goal + +While nothing reads it (the camera is invisible or every encoder is disabled), +`Video.Capture` produces no frames, yet unhiding resumes as quickly as it does +today. + +## Plan + +[moq-dev/moq#7157](https://github.com/moq-dev/moq/pull/7157) made `muted` and +`invisible` keep capture open so unmute and unhide skip getUserMedia. The +capture now keeps producing frames nobody reads. Keep the device open but stop +pulling and processing frames while unread, and measure the idle CPU before and +after. + +Decided 2026-10-30 (maintainer, /quest-plan of the 2026-10-30 /quest-complete follow-ups): ✅ new m1 push-to-talk child / drop. + +Public API: none. Wire: none. diff --git a/quest/m2/README.md b/quest/m2/README.md index eab9e4ad5db..6df26604fe1 100644 --- a/quest/m2/README.md +++ b/quest/m2/README.md @@ -232,6 +232,7 @@ lite-07; bitrate enforcement is not part of this release. - [Clean ALSA playback onset](/quest/m2/audio-alsa-onset.md) - no onset restarts or clicks over PulseAudio - [Interop audit](/quest/m2/audit-interop.md) - our interop runner entries match what we ship, and every red interop cell with a moq.dev side is fixed, reported, or noted as down or unsupported - [The HLS server follows each path once](/quest/m2/hls-follow-once.md) - `Following` reports synced absence, so moq-hls resolves with one follow +- [moq-gst playback shares one session context](/quest/m2/gst-session-context.md) - the playback helpers take one context instead of six arguments - [moq-bench pins the instances it measures](/quest/m2/bench-epoch-pin.md) - pooled requests pin their announced epoch, so a takeover is not counted as the same broadcast - [Bindings see a broadcast end](/quest/m2/binding-broadcast-end.md) - `MoqBroadcastConsumer` exposes the close, and a moq-transport run end is a protocol error - [We own our WebRTC stack](/quest/m2/webrtc/README.md) - moq-rtc becomes moq-webrtc with moq-ice, the data-channel transport, and a vendored, trimmed str0m/dimpl/is/sctp-proto diff --git a/quest/m2/gst-session-context.md b/quest/m2/gst-session-context.md new file mode 100644 index 00000000000..d4699d7fa95 --- /dev/null +++ b/quest/m2/gst-session-context.md @@ -0,0 +1,17 @@ +# [S] moq-gst playback shares one session context + +## Goal + +moq-gst's playback helpers take one session context instead of passing six +arguments of session plumbing to each other. + +## Plan + +Suggested by [moq-dev/moq#7084](https://github.com/moq-dev/moq/pull/7084). +The prerequisite pad-actor refactor landed in +[moq-dev/moq#7368](https://github.com/moq-dev/moq/pull/7368), reshaping the +same code. + +Decided 2026-10-30 (maintainer, /quest-plan of the 2026-10-30 /quest-complete follow-ups): ✅ new m2 quest, after moqsrc-pad-actor / fold into it / drop. + +Public API: none. Wire: none. diff --git a/quest/m2/release/migrate-audit.md b/quest/m2/release/migrate-audit.md index 7ee5d432c3d..74cf1a9a0f7 100644 --- a/quest/m2/release/migrate-audit.md +++ b/quest/m2/release/migrate-audit.md @@ -21,7 +21,15 @@ m1 quest on 2026-10-22. - Correct Unreleased entries that a later PR overturned. Known: `js/net/MIGRATE.md` still describes `request(path, { announced: true })`, which [moq-dev/moq#7173](https://github.com/moq-dev/moq/pull/7173) replaced with - `follow(path)` (decided 2026-10-29). + `follow(path)` (decided 2026-10-29). Also known, from + [moq-dev/moq#7036](https://github.com/moq-dev/moq/pull/7036)'s rename of + `used` to `wanted` (Decided 2026-10-30 (maintainer, /quest-plan of the 2026-10-30 /quest-complete follow-ups)): `rs/moq-json/MIGRATE.md` + should say `producer.demand().wanted().is_some()`, not + `producer.demand().is_used()`. In `js/net/MIGRATE.md`, the line saying + broadcast and group demand keep `used` and `unused()` is false, the line on + `unused()` rejecting on close is moot, and + "`producer.unused()` is `producer.demand().unused()`" should point at + `demand().wanted`. - Run the API-diff report on the cut PR and read the diff of the hand-written binding wrappers (`py/moq-rs`, `kt/moq`, `dart/moq`, `go/wrapper`, `swift`, `cpp/moq`, `cpp/obs`) and of each binary's flags,