Add HIL tests for all peripherals which are practical to test

[jessebraham]

For now, let's use the table from https://github.com/esp-rs/esp-hal/issues/809#issuecomment-1999308797:

I've gathered a list of the esp-hal modules and how hard/easy I think it would be to test those (take it with a grain of salt as I may have misevaluated some). Also, some modules can't be tested, or it does not make sense:

• [x] aes
  • [x] DMA - https://github.com/esp-rs/esp-hal/pull/1426
  • [x] Non- DMA - https://github.com/esp-rs/esp-hal/pull/1297 and https://github.com/esp-rs/esp-hal/pull/1423
• [ ] analog -
• [ ] assist_debug -
• [x] clock -https://github.com/esp-rs/esp-hal/pull/1425
• [x] delay - https://github.com/esp-rs/esp-hal/pull/1415
• [x] ecc - https://github.com/esp-rs/esp-hal/pull/1418
• [x] gpio -
• [ ] hmac -
• [ ] i2c -
• [ ] i2s -
• [ ] lcd_cam -
• [ ] ledc -
• [ ] mcpwm -
• [ ] otg_fs -
• [ ] parl_io -
• [x] pcnt - https://github.com/esp-rs/esp-hal/pull/1746
• [ ] rmt -
• [x] rom - https://github.com/esp-rs/esp-hal/pull/1417
• [x] rsa -https://github.com/esp-rs/esp-hal/pull/1414
• [ ] rtc_cntl -
• [x] sha - https://github.com/esp-rs/esp-hal/pull/1422
• [ ] spi
  • [ ] Half duplex -
  • [x] Full duplex
  • DMA
    • [ ] Half duplex -
    • [x] Full duplex - https://github.com/esp-rs/esp-hal/pull/1443
• [ ] systimer -
• [ ] timer -
• [ ] trace -
• [ ] twai -
• [x] uart
  • [x] blocking
  • [x] async
• [ ] usb_serial_jtag -

[MabezDev]

One reference for the peripherals is the list @SergioGasquez created here: https://github.com/esp-rs/esp-hal/issues/809#issuecomment-1999308797

[SergioGasquez]

I will be working on RSA

[SergioGasquez]

I will be working on CRC

[JurajSadel]

Working on usb-serial-jtag

[SergioGasquez]

Working on SHA

[SergioGasquez]

Working on clock monitoring

[JurajSadel]

I'm currently stuck with usb-serial-jtag, I will work on debug-assist next.

[SergioGasquez]

I will also try to add aes_dma and improve the current aes test to cover all the modes.

[Dominaezzz]

I want to write a HIL test for the LCD_CAM but there's not enough loopback pin connections. At the moment there's only two, any chance I could get two more (at least on the S3)? The alternative is going through the GPIO matrix (which would be ideal I suppose, less faff to run the tests) but that'll need #1662 .

[SergioGasquez]

At the moment there's only two, any chance I could get two more (at least on the S3)?

How many connected pins would you need? Feel free to suggest any pins that you would like to get connected, see https://github.com/esp-rs/esp-hal/issues/1332#issuecomment-2059093479 which might be helpful.

[Dominaezzz]

4 total pairs. A clock pin, a data pin, and two ctrl pins.

According to https://github.com/esp-rs/esp-hal/tree/main/hil-test:

• GPIO2 and GPIO4 are connected.
• GPIO5 and GPIO6 are connected.

I suppose a strategic choice for two more pairs would be:

• GPIO1 and GPIO21. (GPIO1 is the only (excluding strapping pins) pin with the RTC_I2C function).
• GPIO43 and GPIO45. (GPIO43 is U0TXD pin)

My strategy there was to connect a pin with a specific function to another pin with no functions. This is an S3 only choice, as the LCD_CAM only exists there.

[SergioGasquez]

Awesome, I'll try to get those pins connected in the upcoming days. On the ESP32-S3, the following would be added:

• GPIO1 and GPIO21 are connected.
• GPIO43 and GPIO45 are connected.

[SergioGasquez]

@Dominaezzz, our S3 self-hosted runner already has the connections we discussed.