I think I found the reason of my problem:

In ulp_riscv_vectors.S, saving of registers are implemented as:

.equ SAVE_REGS, 17
.equ CONTEXT_SIZE, (SAVE_REGS * 4)

/* Macro which first allocates space on the stack to save general
* purpose registers, and then save them. GP register is excluded ...

This is exactly where the problem is: A function can use s0-s11 for local variables, but when interrupt happens, sx registers are not preserved so when returning from ISR, variable values may change.

In ESP32-S3 RISC-V ULP interrupt code (ulp/ulp_riscv/ulp_core/ulp_riscv_vector.S), there is a note saying:

Note: We don't save the callee-saved s0-s11 registers to save space

This causes some issues that the riscv32-esp-elf-gcc generated code do used s registers for function local variables. If ...

Thanks you're right. I need to manually set my DMM input impedance to 10Gohm to get correct result.

In a low power design I'm putting esp32-s3 in deep sleep and use ext1 io to wake up.

Since the wakeup source is a rotary encoder without dent I can't assume it will stop at open position, so I'd prefer a large pull-up resistor to save power if the wiper stops at close position.

The ESP32-S3 ...

Got it! Thank you.

Hi,

New to ULP program, currently some doubt not so clear:

I'm implementing the ulp firmware as a state machine. For each state, it will decide how long the ULP CPU will sleep and execute the next state, some skeleton code as this:


int main(void) {
switch (s_state) {
case ULP_STATE_INIT ...

No, definitely not using any critical section, I also had interrupt watchdog disabled. The intriguing part is that increase instruction cache or introduce some irrelevant (logging) code within the function call will fix the issue. Is there any other reason level 4 interrupt can trigger other than ...

Hi,
I have a program that runs into a weird situation:

Main component:
ESP32-S3 16MB / No PSRAM, esp-idf 5.1.2, using built-in unity test framework and protobuf-c

Description:
I have 2 test cases covering a protobuf unpack and a protobuf pack. Both test cases run fine individually. But if I run ...

Gd25 3v flash only rated down to 2.7v? So yes it seems 2.3v is not relevant spec.

I think this way too. Seems the datasheet needs update