Repository navigation
Add support for the Metro ESP32-P4 - #11643
Conversation
Include the ADC mapping, stable UID, ROM USB reset, DSI output and shared SDMMC fixes verified during Metro P4 bring-up. Keep board wiring and configuration in a separate dependent commit. Co-authored-by: Limor Fried <ladyada@users.noreply.github.com>
Configure USB-C device, USB-A host, SDIO storage and the C6 Hosted coprocessor with the allocated Adafruit USB ID. Co-authored-by: Limor Fried <ladyada@users.noreply.github.com>
Initialize the required nested runtime after fetching ESP-Hosted, using the same filter or shallow-clone options as mbedtls. Co-authored-by: Limor Fried <ladyada@users.noreply.github.com>
Co-authored-by: Limor Fried <ladyada@users.noreply.github.com>
Retain the existing LDO handle across simulated deep-sleep wakes instead of adding another reference each time board_init runs. Co-authored-by: Limor Fried <ladyada@users.noreply.github.com>
Add Adafruit Metro ESP32-P4 board
# Conflicts: # ports/espressif/Makefile # ports/espressif/common-hal/microcontroller/Processor.c # ports/espressif/common-hal/microcontroller/__init__.c # ports/espressif/common-hal/mipidsi/Display.c # ports/espressif/common-hal/usb_host/Port.c # ports/espressif/esp-idf # ports/espressif/mpconfigport.mk # supervisor/shared/usb/tusb_config.h
ESP-IDF v6.1 integrates Wi-Fi remote support into esp_wifi itself
|
This was fixed for this change: adafruit/esp-idf#21 I don't think it is required though |
dhalbert
left a comment
dhalbert
left a comment
There was a problem hiding this comment.
This looks pretty safe, except that the sdioiochanges could affect other boards. Ideally it would be tested elsewhere.
@mikeysklar do you have a espressif board on your HIL board that has an SD card you could test this against?
|
Yep, I'll try it on a Feather ESP32-S3 with SDIO breakout. |
|
@tannewt all the builds are failling. I did approve and merge the esp-idf change |
|
I'll submit a PR to bump the pin for adafruit/esp-idf. There are several PRs that need this. Testing now to make sure it builds and runs locally. |
|
@mikeysklar Instead of PR, you could push a commit to this PR to fix it, assuming it's clear the new pin will fix the broken builds. |
|
No concern about SDIO regression.
It is running fine on a Feather ESP32-S3 @40 MHz using the SD benchmark |
66cb264d was never pushed, so every espressif build failed at checkout. 29d36039 is the tip of adafruit/esp-idf circuitpython-v6.1 and carries the same SDMMC slot removal guard (adafruit/esp-idf#21). Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Match the esp-idf pin in adafruit#11643 (29d36039) so the two PRs merge cleanly in either order. 29d36039 already contains the jpeg full-range change (adafruit/esp-idf#18) this PR needs, plus #19 and #21. Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
esp-hosted's Kconfig uses $(ESP_IDF_VERSION). export.sh sets it, but the "Set environment" step only carried IDF_PYTHON_ENV_PATH and PATH forward, so ldgen failed on the Metro ESP32-P4 with "macro expanded to blank string". Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
|
The Metro ESP32-P4 CI build failed. ESP_IDF_VERSION was set by export.sh but not passed on to the build step I added it to the CI environment |
dhalbert
left a comment
dhalbert
left a comment
There was a problem hiding this comment.
Thanks @tannewt and thanks @mikeysklar for the fixes!
4 participants
No description provided.