tinyusb.git/test, branch claude/adoring-pasteur-kbaFa

HIL: replace build.flags_on with named build variants (#3687)

2026-06-11T01:16:43Z

* test/hil: replace build.flags_on with named variant schema Boards declare build variants as `variant: [{name, flags}]` instead of `build.flags_on`. The variant `name` is the build dir (cmake-build-) and the HIL report row; `flags` is the raw CFLAGS string (-D...=1) injected via CFLAGS_CLI. No `variant` => a single build named after the board. - build.py: --build-name (dir) + --cflag= (raw CFLAGS, repeatable, =form survives the matrix's shell word-splitting); drop -f1/CFLAGS wrapping. - hil_ci_set_matrix.py: emit one build arg per variant. - hil_test.py: iterate variants; report row + build dir = variant name. - hil_ci.sh: copy all cmake-build-* dirs for -b runs. - get_deps.py: accept (ignore) --build-name/--cflag from matrix args. - tinyusb.json: migrate all 6 flags_on boards to variant. * board_test: park CI build with busy spin instead of wfe

Merge pull request #3690 from hathach/claude/board-test-idle-park

2026-06-10T11:04:54Z

hil: park boards with idle board_test instead of erasing flash

test/hil: erase MCU after tests; show throughput speeds in report

2026-06-09T04:44:06Z

Teardown: instead of flashing device/board_test (a USB-less blink loop that keeps the MCU busy-looping), erase the first flash sector (vector table) so the board faults to idle after its tests — no USB, lower power, faster. Per-flasher erase_: openocd/openocd_adi `flash erase_sector 0 0 0`; stlink `--erase 0`; jlink erases the sector at the flash origin read from the ELF (pure-Python, new elf_flash_origin); esptool `erase_region 0x0 0x4000`; lm4flash writes a 4 KB all-0xFF blank image (lm4flash erases before programming, so the first sector ends up blank). device/board_test flash remains a fallback for flashers with no erase_ function. The teardown is no longer a report column (it's cleanup). Report: cdc_msc_throughput and msc_file_explorer[_freertos] now return a compact read/write speed shown in their report cell instead of the pass tick (e.g. "C 652k/422k M 1.1M/783k", "rd 1.2MB/s"). test_example returns an optional metric; render_matrix shows it verbatim. Firmware lookup factored into find_firmware (reused by the erase teardown). Verified on the rig (stm32f723disco, jlink): erase disables the board in 0.8 s and it disappears from the bus; the throughput cell shows live speeds. Co-Authored-By: Claude Opus 4.8 (1M context)

test/hil: use ⚪ for skipped in HIL report

2026-06-08T10:49:07Z

Neutral white circle for skipped, giving a ✅/❌/⚪ pass/fail/skip set. Co-Authored-By: Claude Opus 4.8 (1M context)

test/hil: use ✅/❌ emoji for HIL report pass/fail

2026-06-08T10:17:26Z

Colored emoji render green/red in the GitHub PR comment, far more visible than the monochrome ✔/✖ dingbats. Skip stays ➖. Co-Authored-By: Claude Opus 4.8 (1M context)

test/hil: clear HIL report up front on a fresh run

2026-06-08T09:18:04Z

The report sidecar lives in a persistent dir (it survives the CI workspace clean so accumulation works across run attempts). A full run is "fresh" and must not merge prior state, but previously fresh only avoided *loading* the json at merge time — if a fresh run crashed before writing the report, the stale json/md from an earlier run lingered and a retry (fresh=False) could merge it, or the always() upload could post it. Delete hil_report.json/.md at the start of a fresh run so prior results can never leak. Co-Authored-By: Claude Opus 4.8 (1M context)

test/hil: accumulate HIL report across re-runs; post as sticky PR comment

2026-06-08T09:11:13Z

hil_test.py persists results in a hil_report.json sidecar and regenerates hil_report.md from it. A full run starts fresh; a re-run (--skip-board / -bt, i.e. the .skip file) merges into the existing report so already-passed boards/tests are preserved while only re-run cells update. The report dir is configurable via HIL_REPORT_DIR. build.yml: each HIL rig writes the report to a workspace-sibling dir that survives the per-attempt workspace clean, and uploads it as an artifact. A new hil-report job merges the rigs' reports into one sticky PR comment (marocchino) with one table per rig. Co-Authored-By: Claude Opus 4.8 (1M context)

presets,hil: keep Ninja Multi-Config; make HIL find its output

2026-06-08T08:42:00Z

Per review (HiFiPhile): Ninja Multi-Config is needed for IAR, otherwise the optimization level can't be lowered to none for debug. Revert gen_presets.py back to Ninja Multi-Config (keeping only the cmake-build- binaryDir change), and instead teach hil_test.py to locate .elf whether it sits directly in the example dir (single-config) or under a per-config subdir like RelWithDebInfo/ (multi-config). Verified: stm32u083nucleo passes 13/13 remote HIL with a multi-config preset build (rsync preserves the RelWithDebInfo/ subdir; the resolver finds it). Co-Authored-By: Claude Opus 4.8 (1M context)

test/hil: report board x test results as a markdown matrix

2026-06-08T08:17:20Z

hil_test.py now writes hil_report.md and prints it to stdout: rows are boards, columns are tests (bare example names), cells are pass/fail/skip. test_example returns a per-test status, test_board collects a board x test grid (one row per flags-on variant), and main() renders an aligned table. A missing binary counts as skipped. hil_ci.sh copies the report back from the remote after a run; hil_report.md is gitignored. Co-Authored-By: Claude Opus 4.8 (1M context)

test/hil: add stm32u083nucleo to pool and expose ~/bin on remote PATH

2026-06-08T07:35:07Z

Add stm32u083nucleo to the active boards in tinyusb.json, flashed via the stlink flasher (onboard ST-Link + STM32CubeProgrammer); ci's openocd build has no STM32U0 flash driver. STM32_Programmer_CLI lives in ~/bin on ci, which the remote `bash -s` shell in hil_ci.sh did not have on PATH, so add $HOME/bin to its PATH export (matching the GHA runner .path). Verified remote: 13/13 device tests pass on ci.lan. Co-Authored-By: Claude Opus 4.8 (1M context)