That was totally it! Thanks!

LOTS of debugging of random crashes - boils down to this:

My Code:

#define RC_CHANNELS 6
unsigned long rc_val[RC_CHANNELS];

When I do:
ESP_LOGE(TAG,"rc_val ptr is at %p size is %d",rc_val,sizeof(rc_val));
As you might expect - it tells me:
E (1972) CouchLED: rc_val ptr is at 0x3ffb2a84 size ...

> Make sure to also de-clockgate and un-reset the RMT peripheral.

@ESP_sprite how do I do that? I added the following - still no interrupts:

#include "driver/rmt.h"
#include "soc/rmt_reg.h"
#include "soc/rmt_struct.h"
#include "esp_intr_alloc.h"

#define RMT_CHANNEL RMT_CHANNEL_2
#define RMT_GPIO ...

I need to write RMT code that doesn't use the RMT driver, and instead bit-bashes and manually sets up GPIOs and interrupts to work. (It has to do with EVENTUALLY needing this to work in conjunction with another piece of code that transmits - and the driver's difficulty to do both at the same time ...

I am working in ESP IDF 4.4.2

I have some off-the-shelf code that I'm working with that do two different things, and both work:

* A library that uses RMT to transmit words to a Neopixel strand
* A library that uses RMT to sniff servo waveforms from an RC remote control receiver

But when I try to ...

I don't want to get TOO deep into the answer here (considering this thread is soooo old) - but the answer is that the Neopixel handling is very timing sensitive - particularly when you have enough pixels that the data being transmitted has to span more than a single RMT buffer. This means that if ...

P.S. It seems as when I set the max duty cycle to 8192 - that made the "fade" work full scale (0% to 100% PWM).

But that actually help reveal that the following code doesn't seem to be doing anything:

printf("3. LEDC set duty = %d without fade\n", LEDC_TEST_DUTY);
for (ch = 0; ch < LEDC_TEST_CH ...

I'm building and running the stock led fade example. It doesn't seem to be working correctly.

Biggest question is the two set-duty examples are supposed to set it to "0" and "test". But in both cases they do 50/50 duty cycle all the time. This is true for BOTH tests.

The other weird issue is that ...

Bumping this ... because several projects are stuck on this. Any idea?

I am having this exact problem - and it is also trying to migrate 4.1 code to 4.2.2

I did what you said here - and there are MANY references to:

-DMBEDTLS_CONFIG_FILE=\\\"mbedtls/esp_config.h\\\ in the compile_commands.json file.

There are are a few contentenders to what this could be in my esp ...