Serial ports in the chooser may belong to other devices (e.g. FT232R),
not the RP2040. Show the BOOTSEL tip whenever no BOOTSEL device is
detected via picotool, regardless of whether serial ports are present.
Replace UF2 file copy to mass storage volume with direct picotool upload.
This avoids macOS "disk not ejected properly" warnings caused by the RP2040
resetting immediately after receiving the firmware.
- Use picotool (already installed by PlatformIO) for BOOTSEL detection
and firmware upload via USB
- Upload ELF directly with `picotool load -v -x` for real-time progress
- Remove platform-specific mass storage volume detection (macOS/Linux/Windows)
- Show helpful udev rules message on Linux permission errors
Move signed bin creation from post_build script to upload_using_platformio.
The post_build AddPostAction only runs during build, not during nobuild
upload, so the file was missing when PlatformIO tried to upload.
Now create firmware.bin.signed before calling PlatformIO upload.
Add auto-detection of RP2040 BOOTSEL mass storage volumes (RPI-RP2) on
macOS, Linux, and Windows. Show detected volumes as upload targets with
a progress bar for UF2 file copy. Display helpful BOOTSEL instructions
when no RP2040 device is found.
- Add get_rp2040_mass_storage_volumes() to detect mounted RPI-RP2 volumes
- Add PortType.MASS_STORAGE and upload_using_uf2_copy() with progress bar
- Move ProgressBar to helpers.py for shared use
- Wait for USB-CDC serial port after upload for log output
- Auto-select single serial port for logs after mass storage upload
- Create firmware.bin.signed in post_build to fix nobuild upload target
- Show BOOTSEL tip when only OTA options are available
Every entity generated two codegen calls:
entity->set_name("Name", hash);
entity->set_entity_strings(packed);
Merge these into a single configure_entity(name, hash, packed) call
to reduce generated code size. For a config with 50+ entities this
eliminates 50+ function calls from the generated setup() function.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Defense-in-depth: validate device class strings don't exceed the
48-byte PROGMEM buffer limit (47 chars + null), matching the same
pattern used for icon strings.
The Python tests were testing a Python reimplementation, not the
actual C++ code. Replace with an integration test that exercises
the real micros_to_millis() on the host platform, covering boundary
values, carry paths, and mod-8 shift edge cases (25 checks).
Make micros_to_millis a template on return type (default uint32_t).
millis_64() now calls micros_to_millis<uint64_t>() to eliminate
__udivdi3 there too — same Euclidean decomposition, just with a
32x32->64 multiply for hi*Q instead of truncating.
Add 64-bit variant tests (140 total, all passing).
Add Python unit tests (70 cases) verifying the Euclidean decomposition
matches us // 1000 across edge cases: small values, hi/lo boundaries,
carry-path overflow, realistic uptimes, and shift-boundary mod-8
variations.
Make micros_to_millis constexpr to enable compile-time validation
via static_assert.
The template wrapper caused the compiler to make different
optimization decisions on setup(), negating the savings.
Calling register_component_ directly was -4 bytes.
Co-Authored-By: J. Nick Koston <nick@koston.org>
Using the original register_component name (just moved to protected)
produces identical compiled output to the baseline, avoiding the
symbol name length overhead from renaming.
Co-Authored-By: J. Nick Koston <nick@koston.org>
J. Nick Koston