GitLab

Added new RawDevice (W6100) for lwIP Ethernet
---------
Co-authored-by: Stefan Nuernberger <stefan@elexir.eu>

* Update to 2.0.0 SDK
* Board type needs to be set before earliest SDK setup
* Platform includes update
* Boot2 files
* Simple compilation issues
* Build and link
* PIO rebuild with version
* Newlib wrapper update
* Force inclusion of runtime_init_* fcns
The linker was dropping all references to the library's included
runtime_init_xxx functions and hence things like the IRQ vector
table and mutexes and alarms weren't properly set up leading to
instant crashes on start up..

Explicitly call out one function from the object file stored in
the .A to force the inclusion of all the functions.  May be a better
way, heppy to hear any ideas.
* Fix SPI GPIO calls
* Fix Ethernet GPIO
* Remove SDK warnings
Remove the skipped error messages once the following PR merged:
https://github.com/raspberrypi/pico-sdk/pull/1786
* BTStack moved SBC encode/decode paths
* Platform.IO fixes
* BT No longer has special absolute mouse
* Rebuild and update OTA
* Rebuild BearSSL, too
* Update liker file to latest SDK
* Clean up libpicocmake
* Clean up LWIP/BT library names

* Add VFS to enable POSIX file I/O operations

Enables use of FILE * operations on internal and external storage.  fopen,
fclose, fseek, fprintf, fscanf, etc. supported.

* Add FS/File::stat and support POSIX stat/fstat

Removed FS::info64, and updates FS::info with the 64-bit version since in
2024 it's almost impossible to get a SD card smaller than 4GB.

Most code can simply replace info64 with info and continue operation, if they
were updated to be 64-bit in the first place.

RP2040::memcpyDMA implements a DMA-controlled memory copy call identical in
function to standard memcpy, but using an onboard DMA engine.  For large
memory transfers this can be significantly faster than using the CPU-based
memcpy.  Only 4-byte aligned source, destination, and counts are allowed.
If any inputs are not 4-byte aligned, then standard memcpy will occur so
it will behave correctly for any inputs.

No functionality differences expected.

The included USB libraries are not compatible with Adafruit TinyUSB, so
add a #error message when they're built with the define set.

Lowered level of segment 'ESP32 Compatibility' in documentation of 'WiFiClientSecure Class' to be a subsection thereof.

Somehow returning the results of a `void` function from another `void`
wrapper didn't trigger any warnings.  Also missed tcp_bind actual
GCC wrapping.

In ported libraries there were still some remaining DEBUG_ESP_PORT
references.  Moved to their RP2040 equivalents.

From @JAndrassy https://github.com/esp8266/Arduino/pull/9173

Create a single spot with the gamepad16's HID report descriptor and
report structure.  Avoids cut-n-pasted code.

Found via #2296, the HTTPClient was looking for old ESP8266 defines and
not the RP2040 core ones to enable it.  Now dump on `Core` level.

Picoprobe was rechristened Debugprobe earlier this year, add a note
in the menus to be specific about it.

See https://github.com/raspberrypi/debugprobe

Underlying HID_Joystick now always using 16-bit format axes, need to update
BT and BLE descriptors sent to the BT master or it will misinterpret the
reports and the reported joystick state will be read as garbage.

Fixes bug introduced in #2276, oops!

Use GH native ones instead.

Variant builds were taking longer than the actual individual CI jobs, so
split it up.

Combine the spelling and style checks, they ran very fast and spent more
time in checking out than in running.

Upstream patch https://github.com/espressif/arduino-esp32/pull/9991

Fixes #2287

Added 4 new boards from Amken LLC.
* Amken Bunny
* Amken Revelop
* Amken Revelop Plus
* Amken Revelop eS
Co-authored-by: H.Keni <151807089+hrken1@users.noreply.github.com>

Fixes #2275

Adds `Joystick.use10bit` and `Joystick.use16bit` methods.  10-bit is
unsigned from 0...1023 while 16-bit is signed -32767..32767.
Defines a new HID descriptor to support the increased resolution.

Co-authored-by: HTY2003 <randumbperson@gmail.com>

The repo and development use only '\n' (UNIX) EOLs.  When a user runs
makeboards on a Windows system they end up changing every line in boards.txt
and the JSON files to Windows '\r\n' format.

Explicitly set the newline character when opening the output files to
avoid this.

* Add a build of all variants to CI using P.IO
* Split out into separate job, use BOOTSEL sketch
* Fix Breadstick variant
* Fix Bridgetech boards with illegal define names
  Dash(-) to underscore(_) in define and variant for the -7 and -43.
* Bridgetech JSON updates
* Temporarily remove Bridgetech boards from CI
  Needs an update to the P.IO external repo to work since the names
  of the boards have changed.

* Fix D pins on Adafruit IB RP2040
* Fix Adafruit Feather D pins mapping as well
* Change common.h to accept D pin def per board

FreeRTOS often seems to have interesting corner cases.  Add two simple
tests that have been useful while debugging issues found from users or
from FreeRTOS updates.

Fixes #2254

The faked certificate was allocated but not deleted in certain cases.  Make sure
to clean up in the destructor.

An oversight in the order of updating the ASFLAGS with a copy of the CCFLAGS (see line 110) and then updating the CCFLAGS (without resyncing the ASFLAGS) leads to a fatal compilation error in the Adafruit PicoDVI library, in which `tmds_encode.S` fails to find the `pico/config.h` include file. This fix updates the ASFLAGS manually after changing the CCFLAGS so that they're equal again, and the library can be compiled.

Fixes #2251

The 2-phase send could get out of whack if transmission was attempted when
no device was connected. Clear things up so if things aren't connected,
then no data gets set as pending.

FreeRTOS has merged the SMP branch into its main, so move to that and
adjust the core accordingly.  V11.1.0 + several minor edits.

* Fixed incorrect AVR compatibility macros.

---------
Co-authored-by: Pontus Oldberg <pontus.oldberg@non.se.com>

- Board information brtchip.com/product/idm2040-43a
Signed-off-by: Tuan Nguyen <tuan.nguyen@brtchip.com>
Co-authored-by: Tuan Nguyen <tuan.nguyen@brtchip.com>

A Timer is not active after the alarm fires once, so clear the
alarm ID so we know we're not running.

Uses the Pico SDK alarms and repeated-timers to provide for IRQ-level
periodic tasks to be scheduled.

Fixes #2231