Docs and technical reference manual inconsistent for clock stretching timeout on ESP32-C3 (IDFGH-9276)

[TheButlah]

Answers checklist.

• [X] I have read the documentation ESP-IDF Programming Guide and the issue is not addressed there.
• [X] I have updated my IDF branch (master or release) to the latest version and checked that the issue is present there.
• [X] I have searched the issue tracker for a similar issue and not found a similar issue.

General issue report

In https://github.com/esp-rs/esp-hal/issues/352 I have been working on delving into esp-idf's implementation of clock stretching.

• i2c_set_timeout()'s docstring states that its delay is specified in APB clock cycles (not exponential).
• It calls i2c_ll_set_tout() which states that the timeout is 2**tout APB clock cycles (exponential). This controls the I2C_TIME_OUT_VALUE parameter of the I2C_TO_REG register (0x0C).
• The technical reference for esp32c3 states that I2C_TIME_OUT_VALUE describes the number of APB clock cycles (not exponential), in a couple of places.

Is I2C_TIME_OUT_VALUE linear or exponential? Is the technical reference correct that its linear?

If it is exponential, the docs for i2c_set_timeout() should be changed, and if its linear, the docs for i2c_ll_set_tout() should be changed.

[TheButlah]

This issue has been marked as done - what was the resolution? @mythbuster5

[mythbuster5]

Yeah.. the docs of i2c_ll_set_tout should be changed. It's linear.

[TheButlah]

We still have contradicting evidence. According to this experiment, i2c_ll_set_tout and I2C_TIME_OUT_VALUE is exponential, and the TRM is wrong.