Page 1 of 1

ULP GPIO problems

Posted: Thu Mar 01, 2018 4:29 pm
by kbaud1
Here is what I found from trial & error:

- RTC bits 0-9 require the following initialization to be used as inputs from the ULP:
rtc_gpio_init (IO#) - I don't see why this is required because the function description says "This function must be called when initializing a pad for an analog function." We are obviously not using the pin for a digital function, not analog.
rtc_gpio_set_direction (IO#, RTC_GPIO_MODE_INPUT_ONLY) - This makes more sense, but you would think it would already default to input like the other pins.
- RTC bits 10-15 do not require any initialization to be used as inputs from the ULP.
- GPIO0 still has a 2.2K pullup resistor that I don't know how to disable.

So even though I figured out how to read all the pins, the logic to configure them doesn't make sense.

Once of the examples I looked at earlier on was:

https://github.com/espressif/esp-idf/bl ... ple_main.c

Which Contains this comment:

/* Initialize GPIO0 as RTC IO, input, disable pullup and pulldown */

This comment is false, since the pullup resistor cannot be disabled? (and is not disabled when this program is run)

Re: ULP GPIO problems

Posted: Fri Mar 02, 2018 1:53 am
by ESP_Sprite
Are you sure the board you use doesn't have an external pull-up on GPIO0? 2.2K sounds too low to be an internal pull-up, these are usually tens of KOhms.

Re: ULP GPIO problems

Posted: Mon Mar 05, 2018 12:35 am
by kbaud1
It is the rev 1 version of the DevkitC.

Re: ULP GPIO problems

Posted: Mon Mar 05, 2018 1:17 am
by ESP_Sprite
Huh, no pull-up there. Can I ask how you measured the 2.2K?

Re: ULP GPIO problems

Posted: Mon Mar 05, 2018 10:50 pm
by kbaud1
I used a 3.3K resistor to pull it down and measured the voltage. The 2.2K was only a rough visual estimate, I did not make a precise measurement or calculation.

Re: ULP GPIO problems

Posted: Mon Mar 05, 2018 11:55 pm
by WiFive
There was some problem with devkitc V1 and the auto reset circuit which could be related?