AM3352: Probability of parity bit error

if there is a parity error on uart, I think it is easier to scope the uart signal like tx/rx first? or you have done it already? can you share more on this?

The detailed introduction about this earlier.

OK, by reading the previous post, you mean:

1. the am335x uart tx is connect to a logic analyzer

2. the am335x runs a program and output the same data (0x85? 85?) in a loop to a uart port

3. the logic analyzer will analyze the received data and label them

4. the logic analyzer find a frame with correct data but wrong parity bit

this still miss some detail like:

1. is the wrong parity happen same pattern every time? I mean every time the data is correct and the parity is wrong.

2. if the data being send the changed like 0x55 or 0xAA, will the pattern different?

3. Can the length of the bits being measured to determine the real baud rate?

4. Can the wrong frame being measured using a scope?

because the above missing detail, I can't make a strong argument, my guess is this is likely to be a baud rate miss match, do you try to change the baud rate a little like Gary suggested?

Firstly, the data sent via UART is cyclic, ranging from 0x00 to 0xFF. The occurrence of errors is random. There was a table in a previous post showing the test results of one of the products.
1. The higher the baud rate, the greater the probability of errors. In this case, the data itself is correct, but there is a probability of error in the parity bit.
2. Every data transmission has a probability of error.
3. Yes, a logic analyzer can calculate the baud rate by measuring the bit length.
4. Oscilloscopes are not ideal for capturing sporadic phenomena, unlike logic analyzers.
5. I have tested our products with AM3352, AM3354, and AM3359, and all of them exhibited this issue. However, when I used the official development board, the problem did not occur (e.g., at a baud rate of 192000, I sent 4800000 bytes without experiencing any isolated parity errors). On the official development board, data becomes scrambled at high baud rates, rather than just having parity errors. Hence, I suspect a hardware issue and would appreciate any suggestions you can provide.

Thank you for your suggestions. I will provide feedback to our hardware team to further investigate the root cause of the issue.

Hello，

Ok, look forward to your feedback, thank you

Regards，

Gary

Hello，

We can now rule out hardware issues. Recently, we expanded our testing scope and found that the code runs fine on our product using the official development board, which only utilizes the bare metal UART functionality. However, we encountered issues when testing our product, which uses multiple peripheral devices. Do you have any software suggestions? Could this problem be caused by peripheral device driver issues or any other factors?

Hello，

1. Check the compatibility of the peripheral device drivers with the AM3352 platform. Make sure that the drivers are up-to-date and properly configured.

2. Ensure that the peripheral devices are properly connected and configured. Check the wiring, power supply, and other relevant parameters.

3. Use debugging tools to identify any potential software issues. This could include tools such as JTAG debuggers or kernel debuggers.

Regards，

Gary