To make me able to accurately study the case, please describe in details and step by step, what exactly i.MX6SoloX hardware do you use (is it the i.MX6SX SABRE SD board or your own custom hardware?), what application software do you run on both A9 and M4 cores, how do you perform the test, what do you expect and what results do you actually get.
In general, the most likely cause of the issue is that the M4 core does not flush its rpmsg receive queue somehow.
Have a great day,
Artur
-------------------------------------------------------------------------------
Note:
- If this post answers your question, please click the "Mark Correct" button. Thank you!
- We are following threads for 7 weeks after the last post, later replies are ignored
Please open a new thread and refer to the closed one, if you have a related question at a later point in time.
-------------------------------------------------------------------------------
Hi, Artur
I have a custom board which copy form I.MX6SX SABRE SD borad,M4 run the "str_echo_bm" demo and modify the data format, from receiving the string to receiving the int value, then the program run by A9 is an infinite loop function, each time sending 100 int values, then M4 prints the first and last values ,when the number of times sent to M4 exceeds 256 times, the values of the first and last outputs of the M4 are different, and some data is similar with the first time.
It still mostly looks like the software issue. Please double-check that the M4 code actually reads all received data from the Message Unit Receive registers on the M4 core side.
Best Regards,
Artur
Hi,Artur
Could you give me any tips ? How can I check Message Unit Receive registers on the M4 core side?
Double-check that each time the M4 side of the Message Unit hardware receives a message, the code, running on the M4 core, explicitly reads it.