tinyusb.git/docs/superpowers/specs, branch claude/usbh-enum-timeout

hil: make hil_report.md a rendering of hil_report.json (#3840)

2026-08-25T08:04:42Z

hil_report.json and hil_report.md were written independently. Four writers produced the markdown and three wrote no JSON at all -- and those three are the paths where a run died, so a JSON consumer saw nothing exactly when it mattered: the per-board verdicts an agent hands back reported the whole fleet as "no report row" while a human read the real story from the markdown. Every writer now goes through render_report(), so a table can never contain something the JSON does not. The document gains `scope` (a three-board PR run and a full run that lost 24 boards were indistinguishable) and `caveat` (how the run ended). `banner` carries rig health across an --accumulate retry; `caveat` records how a run ended and must not -- conflating them made a clean retry publish an abandonment that never happened. helper/hil_report.py owns the document end to end, dissolving the import cycle that forced write_timeout_report to compose its own markdown and removing a duplicate cell classifier kept in sync by hand. hil_summary.py is deleted; its CLI moves there. hil_ci.sh uploads the sidecar so a remote --accumulate has a merge bas

ci_select: fix the membrowse test's env dependence, and stop HIL unit tests taking the rig (#3846)

2026-08-25T03:35:45Z

test_the_upload_board_can_diverge_from_the_built_board called get_family_boards without ci=True, so it pinned the developer's set, not the runner's: the CI skip lists move the one-first pick on three families. It held locally and went red on its first CI run. Pass ci=True, as _prune_buildable already does, and pin the runner's twelve. Rule 2 is a bare test/hil/ prefix, so the harness's own unit tests booked the full 27-board rig for diffs that cannot reach it. Carve test/hil/test/** out to rule 1b, beside test/{fuzz,unit-test}/**; the harness itself is untouched. A test pins that directory's file list, so anything added there that the rig does read fails rather than silently skipping hardware. Rule table updated in the spec and its carbon in the docstring.

ci: an empty selection must build nothing, plus selector follow-ups (#3845)

2026-08-25T02:46:42Z

ci: an empty selection must build nothing, plus selector follow-ups A PR whose build axis legitimately selected nothing rebuilt everything. build.yml reads .build.families twice - as a |-joined regex, and implicitly as "is anything selected" - but tested only -z "$FAMILY_REGEX", which an empty list and a charset-rejected one both satisfy while meaning opposite things. ci_set_matrix had already returned the correct all-empty matrix; the fall-open branch discarded it. #3842 and #3840 each spent 74 cmake legs on it. Branch on the two cases instead, rename FAM_* to FAMILY_*, and cover the block with a test that extracts it from build.yml and executes it - it had no test at all, which is how this shipped through two merges. Follow-ups to the same machinery: glob.escape the repo root at five sites, so a checkout path containing [ or * stops failing closed; drop the ci-full label, read after the matrix was already computed and so never functional; delete 13 mcu:MKL25ZXX / mcu:SAME5X skip tokens matching no board; carry the rule table in the module docstring, guarded against drift; and pin six selection behaviours a mutation pass proved untested. Cut the selector's cost 1.8x (26.0s -> 14.6s) with 0 divergences over 260 paths, and stop scoping the membrowse upload by the PR example filter.

ci_select: classify the 254 files that were reaching rule 17

2026-08-21T15:17:51Z

Rule 17 (unclassified -> full on both axes) is the fail-open net for paths nobody anticipated, and it must stay that way: a wrong `full` costs runner minutes and is visible in the run, a wrong `empty` costs a merged regression and is invisible. But nothing in the tree should REACH it, and 254 tracked files did. The cost was real. PR #3842 changed a skill, a README and .gitignore; .gitignore matched no rule, so both axes went full and 74 cmake legs span up runners to do checkout + toolchain + get_deps before skipping the build, plus the whole 30-board rig. Three changes, none of which touch rule 17 itself: 1. _META_RE - repo metadata and tooling no Build step reads: .gitignore, .gitattributes, .clang-format, .codespellrc, .pre-commit-config.yaml, .readthedocs.yaml, .PVS-Studio/, .idea/, sonar-project.properties, the packaging manifests, CMakePresets, udev rules, test/{fuzz,unit-test} (their own jobs build those), the non-build .github/ files, and the tools/*.py scripts no build invokes. Deliberately NOT included, and still full: .circleci/**, .github/workflows/build*.yml, .github/actions/**, .github/scripts/**. The line is "does a Build step read this", not "is it source". 2. Rules 15 and 16 now match what they already claimed. Row 15 names examples//CMakeLists.txt and the regex never had it; row 16 says tools/build*.py but anchored tools/build\.py$. Both got the right answer only because rule 17 caught them on the way past. Also names their siblings - family_support.mk, family_rules.mk, src/CMakeLists.txt, src/tinyusb.mk - and .circleci/**, which generates the whole CircleCI matrix and was in no row at all. 3. src/typec/** gets row 12b. It is listed unconditionally by both build systems but its body is `#if CFG_TUC_ENABLED`, which only examples/typec/power_delivery sets - the same shape as the class rule, so the same answer: the examples that enable it (stm32g4 and stm32u5 after the buildability prune), and nothing on the rig, which runs no typec test. It was force-fulling 82 families and all 30 boards. TestNoTrackedFileIsUnclassified walks every tracked file and asserts none reaches rule 17, on both axes - 254 -> 0. Verified it fails when a new unclassified path appears. That turns 17 into what it should be: unreachable for anything in the tree, so it fires only for genuinely new shapes, and the author is told to write the row rather than letting the fall-through pick an answer for them. test_full_paths used sonar-project.properties as its stand-in for "unclassified"; that is now metadata, so the case moved to the new test_repo_metadata_is_not_a_build_input, with test_the_build_machinery_is_still_full pinning the other side of the line.

ci: fix nine ways the selection under-selected or mismatched

2026-08-21T05:41:47Z

Every one of these dropped coverage silently - the worst failure mode here, because the PR still goes green. Found by review, each reproduced first. Selection rules: * class_macros derived the config macro from the class DIRECTORY, so a change to src/class/midi/midi2_device.c selected the midi_test examples (which do not compile it) and never examples/device/midi2_device (the only one that enables CFG_TUD_MIDI2, and the only one that does). The file's own macro is unioned in where it differs - union, never replace: over-selecting costs a build, under-selecting merges a break. * the ${FAMILY_MCUS} fallback added for espressif fired on any family whose _family_mcus came back empty, and _cmake_sets is if()-blind and keeps the FIRST definition - so mcx/frdm_mcxn947 answered MCXA15, a token six examples' skip.txt names, dropping 12 firmware images CMake builds. Limited now to families that never spell set(FAMILY_MCUS ...) at all. * lib_examples read only an example's top-level CMakeLists.txt/Makefile; host/msc_file_explorer_freertos names lib/embedded-cli in src/CMakeLists.txt and survived by luck. The whole example tree is scanned. (SEGGER_RTT and rt-thread still resolve to nothing: all three references sit inside a LOGGER=rtt guard no CI build sets - the documented ruling, not a miss.) * get_family_boards applied ci_skip_boards/ci_preferred_boards only under GITHUB_ACTIONS/CIRCLECI, so the selector answered differently on a laptop than on a runner; _prune_buildable forces CI semantics. Its one-board pick also abandoned the whole preferred list when entry one could not build the -e set, and asked skip_example without the build's -D tokens. * _config_enables and lib_examples still read with the locale encoding - under LC_ALL=C the selector tracebacked on three tracked tusb_config.h files. The whole selector and its suite run clean there now. Workflows: * the Membrowse Upload step omitted $EX_ARGS, but --one-first now picks the board from the -e set, so it configured a different, empty build dir and uploaded --identical for a board never compiled. It takes $EX_ARGS for the BOARD; the target stays the aggregate, which has no DEPENDS and still records every example. * blanking FAM_REGEX reset only build_filtered, leaving the build scoped while code-metrics took the UNSCOPED branch and diffed a 1-family run against the full averaged baseline. All three drop together now, as CircleCI's fall-open does. * CircleCI's EX_ARGS had no character screen and is used unquoted, and its code-metrics job still exit 1'd on an empty metrics set - which a scoped build makes a legitimate outcome. * a `ci-full` PR label now turns the scoping off for one PR. A selector bug under-selects silently, and without a label the only ways back to a full matrix are accidental. Performance, since the selector gates every other job: family.cmake texts are read once rather than per changed directory (a 6,000-file dep bump re-read 84 files 99,892 times) and _scrape_mcu is cached: 2.2s -> 0.29s there, 0.8s -> 0.33s on a class diff. Tests: a drift guard for hw/bsp families absent from ci_set_matrix.family_list (they select zero legs now, where they used to ride the full matrix); the rule-4 port test asserted a SUBSET, which set() satisfies, so it could not fail on the empty selection it exists to catch; the GITHUB_ENV guard test counted a SUM of two guards. Drops metrics.py's --only-examples, which nothing called, and applies the TOTAL scrub to the by-example branch that skipped it.

docs: record the CI selection design and its plan

2026-08-21T04:09:03Z

The binding rule table (17 rows x 3 answer columns), the measured effect per PR shape, and the reasoning behind the parts that look surprising: why empty means empty, why hw/mcu and lib are rules rather than full-matrix paths, why get_deps.py is diffed as data, and which build system is the reference. The plan is the task-by-task record of how it was built, kept as the origin trail.

docs: spec, plan and outcome of the .claude instruction-surface audit

2026-08-20T09:47:23Z

Nothing checked the agents, workflows and skills against the code they describe, and the surface had drifted into stating incompatible rules. This records the protocol that found the defects and what it cost. Method: parallel subagents extract every falsifiable claim into JSONL ledgers; a validator re-reads each cited line and rejects any ledger whose quoted text is not there, so an extractor that paraphrases or hallucinates fails a script rather than reaching the verification queue - 1,387 claims, zero such failures. Verification runs only in the main session, and the same gate pointed at `citation` then checks the verifier's own work. Hard-earned evidence is source of truth: code is authoritative about code, experience about hardware, so claims get a fourth verdict, EARNED, and "no backing found" is never grounds for deletion. All 1,387 claims carry a verdict; the behavior sweep deliberately never emits CONFIRMED from a token match, because finding a claim's vocabulary proves presence, not truth. Every real defect came from cross-document comparison - none from any mechanical pass. A path-existence lint was built, measured (11 flags on the audited tree, all false positives, and the target defect is lexically identical to correct text elsewhere), and rejected; recorded so nobody rebuilds it.

examples: document and work around the i.MX RT and LPC55 USB errata

2026-08-18T15:07:49Z

ERR050101: while an isochronous IN endpoint is active, an IN token addressed to that same endpoint number on ANOTHER device sharing the host can silently unprime one of this device's OUT endpoints - control, bulk, interrupt or isochronous alike. NXP states it cannot be detected by software and raises no interrupt, so the endpoint simply stops answering and the transfer never completes. The workaround is a uniqueness requirement rather than a particular number: the isochronous IN endpoint must not share its number with any IN endpoint in use on the bus. One family-wide constant therefore defeats it, since two affected boards on the same hub then pick the same number and each becomes the other's aggressor. CFG_TUSB_MIMXRT1XXX_ERRATA_ERR050101 is set only for the parts whose errata list it - RT1015, RT1020, RT1024 and RT1050, where it is marked no fix scheduled, plus RT1060 and RT1064 rev A - so RT1010 and the RT11xx family keep the ordinary number and cannot collide with an affected board beside them. Several affected boards on one hub can still be given distinct numbers with -DEPNUM_ISO_IN. The guard covers every example that has an isochronous IN endpoint: audio_test, audio_4_channel_mic, uac2_headset, cdc_uac2, usbtest, video_capture and video_capture_2ch. The video examples move the endpoint only when streaming isochronously, since the bulk configuration is unaffected, and video_capture_2ch takes two numbers because it has two streams. The macro name follows CFG_TUSB_RP2_ERRATA_E2/E4/E15 already in tree, and its is fixed, and which cannot be told apart at compile time - a way to define it to 0. device_issues.rst records ERR050101 against every affected part with a link to each errata sheet, and adds the LPC55S2x USB.3 speed-detection and USB.5 isochronous IN entries, neither of which TinyUSB works around. The branch's design notes are included under docs/superpowers. Verified: 340 wedge-free runs on mimxrt1064_evk, which previously wedged within hours, and the macro resolving to endpoint 0x87 on mimxrt1064_evk against 0x83 on mimxrt1010_evk and stm32f407disco.

skills, docs: rewrite USB recovery from the live incidents

2026-08-18T05:19:09Z

Two things the rig taught us that the old guidance got wrong. A usbfs ioctl wedged in D state cannot be freed on a running kernel. It holds the device lock, so usb_disconnect() blocks behind it; reboot(2) walks device_shutdown() and takes the same lock, so every userspace reboot stalls too. Only sysrq b (emergency_restart, which skips device_shutdown) or hypervisor action clears it -- all cited to the kernel source. The recovery ladder is generic across rigs now (ci.lan, hifiphile, a bench PC) and ends at hypervisor escalation only where host access exists. Two claims are corrected outright: JLinkExe is NOT convoy-safe, and a park-flash cannot free a device-lock owner. The hil skill's banner list is what an operator agent matches a report against, so it enumerates the banners that actually exist, including the D-state note -- which is explicitly NOT a wedge, since a healthy in-flight testusb is uninterruptible for most of every case and a concurrent CI battery would otherwise turn a clean run red.

test/hil, ci: contain a wedged USB stack instead of stranding the runner

2026-08-18T05:19:09Z

A wedged USB device used to take the whole HIL run with it. Every worker that touched the poisoned node blocked uninterruptibly, the pool could not be joined, map_async discarded every board's result, and the job ran to the GitHub ceiling with no report at all -- while the self-hosted runner's single job slot stayed occupied and every queued job waited behind it. Bound the calls a worker makes itself. read_sysfs, bounded_open and run_cmd all answer within a wall clock; read_sysfs distinguishes "absent" from "unknown", because a blocked read is not evidence of absence, and caps stranded readers at four (each costs a thread and an fd for the life of the process) after which the worker declares itself blind. mtype, the gio unmount, the libmtp session and the arecord/iperf reaps go through those bounds; the MTP session runs in a disposable subprocess, since libmtp's ctypes calls block unkillably in D state. Bound the run. A pool guard (HIL_POOL_TIMEOUT, 60 min) fires before any job ceiling and still writes a report. When the pool will not shut down, the sweep kills what the workers spawned -- descendants, not just direct children, since flashers run in their own session -- confirms each kill actually landed, and exits early so the runner is freed. Whatever survived is named in the report. Deliberately shallow past that point. We do not re-scan process groups, prove pid ownership, or escalate through sudo: a root-owned survivor is reported, not force-killed, because signalling a pid we cannot prove is ours is the worse failure, and the job ceiling backstops whatever this misses. A D-state holder was never killable anyway. Recover instead of reporting a wedge. A HUNG usbtest case reflashes its own DUT through its roster flasher, but only where the flasher can reach its probe past a poisoned node -- openocd pinned to a validated vid_pid, or esptool. Where it cannot, the run says so rather than reserving budget for a path that cannot fire. Raise the CI ceilings above the pool guard so the guard fires first and still writes its report, and pin --retry 1 on every HIL leg: the guard is a flat constant and does not scale with max_retry, so argparse's default of 3 would triple the serialized usbtest tail against an unchanged guard. Split the module: execution in hil_test/hil_flash/usbtest, infrastructure in helper/ (locking, health, selection, shared bounded IO), and the two matrix generators into .github/scripts/ -- ci_set_matrix.py sat in workflows/, where GitHub treats every file as a workflow definition. 193 tests cover the bounded paths, the kill ladder, the guard and the selector against synthetic /proc trees and PATH-injected fakes; a real wedge cannot be manufactured on demand.